💬┊general

[hasty dagger]

Im Justin from Pittsburgh and I work with a team of embedded developers at Powersoft19

[soft plover]

oh nice! Feels glad to know about it

[hexed verge]

Hi, I am Hameed from Pakistan

[hasty dagger]

they always tell me the weather is hot, how is it there now?

[mild wadi]

Where can I find the white paper thing from the stream yesterday

[hasty dagger]

https://www.kaggle.com/whitepaper-foundational-llm-and-text-generation

[mild wadi]

Thank You!

[pearl pecan]

hey Kaggle community my name's Isaac and I'm joining in from NYC. Looking forward to learning more about LLM's and how they work, and what Google has to offer through this program 🙂

[livid glade]

Hi everyone 👋

[balmy geyser]

hi this is pranav from india just started with ml and enrolled myself for the gen ai course

[cold spade]

Hi this is David from Mexico

[crystal shale]

https://www.kaggle.com/code/nikolasgegenava/license-text-recognition-detection

Hello! Check our my new notebook and upvote if you dont mind 🙂

[cobalt tiger]

hi

[rose dirge]

Hello. I'm Mathew, very pleased to be here.

[orchid trout]

Hi everyone, Glad to meet and hope will help each other to learn together

[nova merlin]

Fun fact: ur name is very similar to deepseek 💀

[tulip igloo]

Hello everone!

I am newly joined member in this discord

[wicked berry]

Hello everyone! Harshada here. Nice to meet you all.

[heady rune]

Viva México! 🙂

[elfin sky]

Hello everyone! I'm Andrés from Colombia.

[cloud echo]

Hello everyone! I am excited to be here. Looking forward to learning with you

[ivory saddle]

Hello everyone! I'm Shivam from India 👋

[covert kite]

Hi everyone

[low lichen]

Hello folks! I'm from India, It is good to be here and looking forward to learn more and exploring opportunities with everyone

[rigid axle]

Hello everyone! I'm Edison from Hong Kong.

[tepid gale]

Hello Everyone, from Southern California

[rancid owl]

hello everyone

[zealous basin]

hello

[night pilot]

This Ahmadu Abdulmalik from Nigeria very happy to be here

[hot orchid]

Hello, I'm from Argentina, very happy to be here

[midnight urchin]

Hello i'm Marie and come from Bénin happy to be here

[pale crystal]

Hello, I am Bader from Melbourne

[sacred mason]

Cheers mates, running a biot late and got lots of work to catch up!

[wide crane]

Hello I'm Jonathan, and I'm completely new to the kaggle platform. I'm a he

I've been trying to verify my kaggle account via phone number but to no avail. I've tried over and over again sending code to my number but still no code shows up. Please, any way around this?

[glass rose]

Hi, My name is Manuel Rodriguez from San Juan, Puerto Rico, and I'm ready to learn and expand my horizons.

[edgy siren]

Hey everyone, Thejas here!

[copper sapphire]

hello

[languid oyster]

Hello Team! I am very excited to join Kaggle.

[languid oyster]

I just came across this page https://www.kaggle.com/code/markishere/day-0-troubleshooting-and-faqs/notebook and the section Error. they are discussing the phone verification error.

[sturdy geyser]

hello

[nova merlin]

hello

[tight frigate]

Hello

[devout sigil]

Hello everyone, Usman here!

[foggy oxide]

Hi! Vinay here

[brave oak]

hello

[nimble knoll]

Hello anyone here online ?

And completed day 3 assignment?

[frosty raptor]

Hello, this is Raenbula from Sincelejo, Colombia

[snow saffron]

if the real abhishek is here then i am a success

[timid idol]

Hi everyone !

[whole gate]

my name is Micheal, i am s data scientist, all look forward to collaborate and build ML models to solve real life problems

[supple zephyr]

Hello All, This is Vineeth Kumar from Michigan. Lets enjoy Learning together.

[sacred flint]

Hi everyone, I'm Saikat. Glad to join this amazing community.....lets learn & grow together.

[fleet quartz]

Hi everyone, I'm Eyasen. Glad to meet, and hope will help each other to learn together 😉

[primal solstice]

Did you guys just hear about the new open weight announced by open ai ?!?

[ripe veldt]

Hi Everyone, I'm @ripe veldt
lad to join this amazing community.....lets learn & grow together.

[real citrus]

Hi, I'm Emmanuel, a Software Engineer from Nigeria. Interested in learning more about Machine Learning and building great projects.

[boreal tiger]

Hi I am Brill from Ghana and it's a pleasure to know you all

[dim crag]

hi everyone did anyone have a problem with kaggle phone verification if so what did you do?

[tall stag]

hello evryone

[still storm]

Hello everyone. shweta here . Nice to meet you all

[simple python]

Hi everyone 👋

[astral badger]

@everyone Hi

[clear wind]

Proven leader in digital transformation, specializing in the strategic implementation of AI/ML, cloud, and emerging technologies including Generative AI and Agentic AI. Empowering organizations to leverage these advancements for impactful business outcomes at Onward Tech as Vice President Delivery.

[lilac yoke]

UPDATE: {RKaggle} is on CRAN 🥳

https://cran.r-project.org/web/packages/RKaggle/index.html

[coral yoke]

Hello!

[somber stag]

Hello

[strong wren]

Hi everyone! I am Pravallika from India

[heavy smelt]

Hie Kagglers. I am Nerissa. New to this but excited to learn!

[fringe notch]

Hi everyone. This is Haris Shah join today 😔. Need your guidance and support to participate

[agile cave]

coding

[copper sapphire]

hello

[agile cave]

What Music do y'all listen to, while Coding? I have a preference for Jungle & Drum and Bass, myself, https://youtu.be/DG5_gQZJPRg?list=RDEM80bYcD6AnOqsnEYaFfqV0w

[vestal crown]

Why does the 5dgai have to trigger a everyone ping?

[agile cave]

I don't mind it. It's something for entire Community here to (potentially) become part of, and notifying everyone of the opportunity.

[umbral heron]

Oop, server guide says to update bio, add kaggle url, and pronouns but drops us in here when you click it lol.

[agile cave]

I prefer to be identified by Adjectives and Adverbs VS Pronouns

[umbral heron]

You're quite extraordinary . Just stating how I got to this channel and what the discord server guide had said. You do you and add your adjectives and adverbs .

[agile cave]

So how did you get into AI/ML?

[umbral heron]

Feeling the need to stay competitive in the workforce lol.

[agile cave]

Makes $ense

[wheat glade]

Hello Everyone,

I am from Austin, Tx 🇺🇸
I am a developer and PhD researcher in AI/ML with cloud/security.

I look forward to connecting with like-minded people.
https://www.linkedin.com/in/akshaymittal143/

[mortal plover]

Hello Kagglers and 5DGAI people. I'm a product manager getting into learning AI.

[gray lantern]

Hey! I’m Jesus, from Mexico and I’m an enthusiast, mainly in electrical engineering concepts.

[mental grail]

Need netflix account for dmc anime

[cerulean trellis]

[nova merlin]

Wrong server

[candid ridge]

I there am Lungile here

[mental grail]

where right server?

[rugged night]

Hi Pratik Manghwani here

[nova merlin]

Sorry i replied wrong person

[silk spear]

Hi there. Hsabul here in Canada

[timber pilot]

Hi, I'm Hriday from Bangladesh

[eternal horizon]

Hey! I am Priyanshu Goswami from India

[outer geode]

Any students out here, Ignore this message if you already have perplexity pro, if not then, sign up with your student email id and get 1 year + 1 month of Perplexity Pro for free https://plex.it/referrals/QLS4KHXF For those who don't know what perplexity is : It is an AI search Engine competing with Google and provides you access to latest llm models all at one place, also you can do deep research and generate images

[fallen turret]

hi

[digital finch]

I like soft music in general

[nova merlin]

I don't code anyways lofi

[agile ferry]

Any crazy people who wants to build crazy stuff!!
Dm me we will figure it out and build somethin crazy!!

[snow saffron]

you just said I

[frigid gorge]

Will they provide certificate for this?

[ivory flicker]

Hey, I'm Lamia. Glad to be part of a great learning community!

[red field]

hi. i am zafar. happy to share knowledge with others

[spring nova]

Hey, I am Ozan from Istanbul, Turkey. He/him if that's asked too. Have fun and learn respobsibly everybody:P

[jolly kelp]

hi i'm Alex. glad to join this course

[echo lantern]

Hello This is Annie from Canada

[blissful latch]

Hi Im Tuan so glad to be part of this AI community

[burnt thistle]

Hlo This Is Vittal Rao From India

[vivid lava]

Hello Everyone,
Where do I mark the assignments done?
Also, where can we get the capstone?

[craggy trail]

Hello Everyone,
I'm from Nigeria and I'll like to be part of a team for the capstone project

[umbral pendant]

Hello fam!! I'm Sheikh Ali living in Islamabad, Pakistan . I'm a Ai Graduate from SZABIST university.

[lament lynx]

Hello from A.P, India.

[umbral pendant]

You can verify with your email address

[craggy trail]

Hello from Nigeria

[heavy steppe]

Hi I'm from India. Glad to be here.

[knotty void]

Hi from egypt,
I have a question. is there any sections befor the daily assignments?.

[twin garden]

Hi, I'm from Singapore. Glad to be here.

[orchid patio]

Hi, I'm from India.

[worthy narwhal]

Hi, I'm Jino from India

[frank kraken]

Hello there

[weak lava]

HI, I'm Aishwarya.

[naive mountain]

Hi, any aspiring Data Analyst here?

[vestal epoch]

hi im kucingbetina from indonesia

[dull root]

Hy im Yashpalsinh from 🇮🇳 and i love PyTorch & Deep Learning

[prisma isle]

me

[opaque basin]

Hi I'm Md Seam Ali Ahammod
Happy to learn with you.
here's my LinkedIn profile: https://www.linkedin.com/in/md-seam-ali-ahammod-pramaniq-8895861a1/

[dim yew]

Great to connect with you all from around the world. Appreciate you Kaggle!

[digital sparrow]

hey

[fast imp]

hi, I'm sannala akash from india.

[modest badger]

Hello Im Gautam from India , and looking forward for making new friends and keen on collaborating

[fast imp]

Can anyone tell what to do in assignments?? any problem statement are posted to solve ??

[past swift]

hey there, My name is Mahnoor

[south warren]

if anyone from TN dm me

[dapper linden]

Hey

[dreamy ember]

Hi I am Shiwansh from India, looking for collabrations

[plush tide]

Hi, from Ecuador South America. It has been great to learn through this wonderful week! Hope to reach some of you in the near future!

[midnight karma]

Hi I am likhith from india

[shadow glen]

Hello

[cerulean trellis]

Hi I am Keyur from Gujarat

[clever hemlock]

hey

[carmine canopy]

Hey I am Dhwani from India . Glad to be part of a great learning community!

[sleek badger]

Hi I am Stuti

[teal notch]

Hi I'm Ebuka from Nigeria

[open nova]

what is platform to upload my project create the job portfolio

[fringe meadow]

Hi I'm T Mohamed Yaser
Happy to learn with you.
here's my LinkedIn profile:
https://www.linkedin.com/feed/update/urn:li:ugcPost:7314304150522011648/

[hoary linden]

Hi I'm Jack (Mechatronics Engineering Senior Student.)

[mint sky]

Hi I am Lakpa

[quiet field]

Hey I am Sneha from India!! (Masters in Operational Research student).

[humble notch]

Hi I'm Jonatas from Brazil

[keen path]

Hi I am Braylin from America

[nova merlin]

Genai made it so smooth

[tardy osprey]

Hey everyone!
I’m working on a data science project using Streamlit and yfinance to build a web app for financial prediction (time series based). Looking for a project partner to collaborate and learn together. Beginners are welcome! Let’s build something cool — DM if interested.

[nova merlin]

Anyone ve experience with singlestore notebooks ?

[violet fog]

Hey guys, I am Ekansh Jindal from Bengaluru, India. Looking forward to learn from you all as I am completely new to this

[lofty matrix]

Hi , I am Nadjib , data science student in Algeria

[upbeat tangle]

Hi! I'm Nhat Minh, a high school senior (getting into college this fall) and passionate about genAI and I have learnt a lot of new things. I have learnt a lot about genAI since October 2024 and I try to start some projects. I'm excited to connect with others who share an interest in AI development:
https://www.linkedin.com/in/nhat-minh-tran-ngoc-1b90642a9/

I also will study in the United States or Canada in the next four years so I would like to connect with you to do many of the significant projects!

[timber scaffold]

Hi! I'm Himanshi , data science and ML student , keen to contribute, from India

[tardy osprey]

Nice to meet you

[tardy osprey]

Have you ever complete any type of project?

[timber scaffold]

yep in CV and NLP

[opal shard]

I don't understand your education! What do you mainly study in Ops research?

[tardy osprey]

So what's your current plan?

And your education?

[timber scaffold]

learning and working on projects as much as I can

[timber scaffold]

Btech CSE, started learning ML from mid of 3rd year

[tardy osprey]

Do you have interest in time series project?

In finance field

I think ml is also includes

[timber scaffold]

Yes

[tardy osprey]

If you want to do this project together feel free to reach out me through my dm

[quiet field]

It's a course where study stats, Cs and Operational Research all three together..

[indigo dock]

I keep forgetting the syntax; is that normal?

[fallow grove]

hi, hope to learn something fun and meet some people interesting

[fierce bough]

I am not sure if someone has already brought this up to Kaggle.

For the Capstone 2025Q1, please fix the number of team members. In the overview tab/section, it' written that a team must have at most four (4) participants. On the other hand, in the rules tab/section, the maximum team size is written five (5).

[wicked berry]

hello

[autumn cliff]

Hello! I hope I can learn about AI and get help in solving Kaggle code problems

[south gulch]

Hey guys
can someone review my resume?
it would be great help

[crystal shale]

https://www.kaggle.com/code/nikolasgegenava/air-pollution-100-accuracy?scriptVersionId=232513032 Check out new interesting solution on air pollution synthethic dataset. Best accuracy before my notebook: 40% :). Upvote if you don't mind

[quiet ginkgo]

🚨 Final Call – Capstone Team Form 🚨

Hey everyone!
This is the last reminder to fill out the form if you're interested in joining my Capstone project team focused on AI/ML-based real-world solutions.

🗓️ Deadline: April 9, 2025
📄 Form Link: https://docs.google.com/forms/d/1UKhtfdgH8Z1CkGXXRIsQ2uaDYC5WBJiXhshn6fj3U4g

Looking forward to connecting with like-minded folks!
📧 Reach out if you have questions: [e-mail]

[indigo dock]

Hey everyone, is it fine if I forget the syntax? I totally get the logic and what to do next, just the syntax slips my mind.

[stuck pecan]

Which method is best for image identification ml in terms of accuracy?

[nova merlin]

ViTs

[opal shard]

Do you mean the study on the workflow of research by "Operational Research"?

[candid thunder]

Kaggle hasnt been working on my browser for now 7 hours

Anyone facing an issue?

[deep cove]

Hi, I just shared a post about my job search. I’d really appreciate it if you could give it a quick like or comment to help boost visibility. Thanks.
Here’s the link: https://www.linkedin.com/posts/mayurkishorkumar_opt-stemopt-opentowork-activity-7315433777474781184-2YIY

[fervent ravine]

I want to become a ML engineer can anyone please guide me
right now i just done my matric exams and i am free for 3 months i want to start learning in this time period
so from where should i start and how i should continue my study

[bright orchid]

Is is possible to find the chat message in closed matches before ?

[river pagoda]

i have many time trying to find people that have the same pasion about data analyst and ai like me, i´m training guy and i fell a little lost with the stadistic topics

if someone wants talk with me about this entire world, do it

[rich yew]

Is anybody got any type of certificate for completion

[viscid hollow]

Hello everyone, I'm Sai Swaroop ,from India and a recent graduate from IIT Bombay. I'm currently working in a data science role and looking to expand my skill set in both data science and data engineering. If anyone is interested in collaborating and growing together, feel free to DM me

[vague merlin]

Hey y'all. Tsup! Norman here. On the road to becoming a proficient ML engineer. Excited!!

[half yoke]

Hey

[peak gorge]

distributed at the end of April i believe

[tame mesa]

Hello

Can I contact support

[scarlet walrus]

Greetings friends! Can someone be so kind as to answer me something about Google having updated the Google AI Studio interface in full competition? I lost access to the model I trained...

[orchid cloak]

hello

[vale pine]

https://www.kaggle.com/code/bhaskarmishra44796/video-game-sale-prediction
Can anyone rate my ml code? Its my first ML code

[shadow glen]

Hello @everyone

I'm looking for a web developer

Dm me guys if anyone willing

[nova merlin]

30$/hr

[shadow glen]

Bruh

[proper linden]

Is Kaggle down?

It’s been close to 4 hours and it’s not working on my browser.

[royal crystal]

Hey everyone!
I'm a 3rd-year student and a beginner in the world of Data Science, AI, and Machine Learning. I'm really excited to learn and grow in these fields, and I'm here to connect with like-minded people who are also exploring or passionate about the same areas.
Looking forward to learning together and sharing ideas!

[digital finch]

@rich halo

[gilded crater]

Hi anyone from Dallas ?

[uncut dove]

can someone help me with the capstone
where do i write the code

[nova merlin]

Too many cats

[hazy oriole]

hello

[grave wharf]

https://www.linkedin.com/posts/anamikaghosh18_scselpu-datascience-electiondata-activity-7316146755115307011-VN9m?utm_source=share&utm_medium=member_android&rcm=ACoAAEVOykwBjO_frt00gqnmXNSezb0eGaN8QjQ

Guys do like and comment 👍💬 and connect with me on LinkedIn

[nova merlin]

-# why LinkedIn not X

[dire halo]

If you want to review, comment and upvote my notebooks, I would appreciate it.

https://www.kaggle.com/iskorpittt/code

[urban beacon]

yupsie

[indigo dock]

Ohk, thanks ^^

[crystal shale]

https://www.kaggle.com/code/nikolasgegenava/pneumonia-prediction-baseline

Hello. Check out the new computer vision notebook: Pneumonia Prediction On Xray Images, 95% Accuracy. Pipeline is made, which can be used to train several models on the CV dataset, and evaluating performance for each of them! This pipeline might help choose which model is best for your dataset.

[digital finch]

Very cool

How's ioai prep going btw 😃

[crystal shale]

Perfect! There will be national olympiad in 2 weeks. What about you?😁

[digital finch]

2 stages have completed third stage is going on 😃

[crystal shale]

Have you succeeded in those two??

[nova merlin]

-# ofc he did

-# if he couldn't then no one else could

-# hes like benchmark

[digital finch]

Thankfully yes 😃

All the best for you too

[crystal shale]

All the best for you too! Good luck, Hope we meet in Beijing 🤝🏻❤️

[half yoke]

woww

our olympiad its in 2 weeks from now but the county heat

not the national

[crystal shale]

Good luck! Is it practical or theory?

[half yoke]

county and national its practical

the selection for the national team its both

it's only the first edition tho

[crystal shale]

Perfect, good luck on the olympiad 🌟

[half yoke]

Good luck to you too!

[nova merlin]

-# everyone participating first time do leave Singaporeans n polish guys behind this time

-# atleast individually u can

-# all the very best

[rich halo]

harmonious_quokka_69353_44153, you need to wait 3 seconds to use that command again.

[rich halo]

A cute fluffy cat - @scarlet gull (85%) (slow)

We're very sorry, but something went wrong with your generation. Please try again with a new prompt!

[digital finch]

wow which country are you from if you do not mind?

[half yoke]

romania

wbu

[nova merlin]

India

[stable patrol]

is there any turkish people?? please dm me

[digital finch]

im fro India

[digital finch]

IOAI?

[half yoke]

do u have ai olympiad there?

[digital finch]

yeah first season

[half yoke]

same

[nova merlin]

Bro how rareshika already did participate from ur country last season

[half yoke]

it was closed

[nova merlin]

So it's not first season in ur country

[half yoke]

it's the first edition

it was not an official olympiad last year

[nova merlin]

Wdym romania did sent teams last year at ioai 💀

[half yoke]

only a selection for math medalist

exactly it was a closed selection

[nova merlin]

-# they js didn't recieve prizes

[half yoke]

they did lmao

silver and bronze

[nova merlin]

It was like they didn't let a guy participate in both IOAI n iaio

[half yoke]

yea and not anycone could like be selected

u had

to be

a math olympian

or in informatics

this is the first official year in which NOAI it's open

county heat starts 29th

[nova merlin]

-# stop it go grind or u won't even get bronze see qafig he's eyeing gold

[hasty spoke]

Hi

[digital finch]

I have friends from Romania

for them AI olympiad was only accessible if you came from some other olympiad

[nova merlin]

For all nations it was the same last year

-# but I don't wanna talk about IOAI anymore anyways 🤧

[nimble matrix]

hi

[wind crypt]

Hey everyone! I’m working on my Final Year Project for BS AI and looking for some good, practical ideas. I’m interested in things like computer vision, NLP, generative AI, or anything that solves a real-world problem but isn’t too complicated to build. If you’ve got any suggestions or have seen something cool recently, I’d really love to hear your thoughts!

[nova merlin]

Real world problem n not too complicated both r contrary to each other tho

U might give Vesuvius challenge a try

[heavy depot]

Hey all, can I ask a question?

[vale pine]

[frank cloak]

Wow how did that happen

[vale pine]

Don't know

[frank cloak]

Did you normalize and scale your sets in different ways?

or somewhere an extra square or smth 💀

[vale pine]

I did everything, but it seems like the data is insufficient to make any decent prediction. I even asked ChatGPT to optimize it, but the code it gave me had -29% accuracy.

[frank cloak]

How much data you got?

well something has to go wrong somewhere other than the datasize to give you a negative prediction, even tho i think thats just a chatgpt thing

[vale pine]

https://www.kaggle.com/code/bhaskarmishra44796/2-pred-wtf

[vale pine]

This is the code

[frank cloak]

you have great emotes

[river harness]

please visit my github and follow me
https://github.com/sajjadahmad-dev

[vale pine]

These are discord default emotes

[bitter bone]

This morning I decided to change the model a little.

[half yoke]

this type of performance needs to be awarded a gold medal ngl

[unique yarrow]

Good morning guys, I'm not entire sure where this goes. My teammate and I are working on the Google AI Capstone project and we're running into issues with our code. We're using Gradio to make a graphical interface and while it works, if I edit the code on my end and update our notebook, my teammate can't see the changes. I'm using Firefox and he's using Brave, so we're thinking it could be a browser issue, but I wanted to ask and see if any of you had come across this before and how you solved it. Thank you!

[nova merlin]

Shift to streamlit it's js better

[unique yarrow]

Unfortunately, it looks like Kaggle doesn't support that and our project has to be done in the Kaggle notebook.

[nova merlin]

If u can use gradio u can use streamlit too ig

Besides gradio n streamlit are mostly for testing/demo purposes

[unique yarrow]

Okay, cool.

[vale pine]

How to get this score?

[vale pine]

Nvm i figured it out

[formal tendon]

There's a dog running back and fort in my kernel, can I make it into a cat

[golden turtle]

anyone up or new here ?

[drowsy granite]

Me

[nova merlin]

U r an al/ml engineer right 💀

[river harness]

🔗 Connect & collaborate with me on GitHub: https://github.com/sajjadahmad-dev
💼 Also on LinkedIn https://www.linkedin.com/in/sajjadahmad-dev/

[drowsy granite]

Yes

[neon zinc]

Hi! Do anyone know why a Notebook appears like this in Kaggle? When editing, it appears correctly and well rendered. But after saving it, it appears all unformatted, like in the image. I can't figure out why... 😦

[flat tendon]

does it still appear like this even after refreshing?

[neon zinc]

Yes... 😦

[flat tendon]

see if this helps

[neon zinc]

Nop, I tried to change the CSS rules in the inspector of the browser.

And I even tried to add a HTML render cell like

# Import HTML from IPython's Core Display module
from IPython.core.display import HTML


# Render a CSS style script for HTML
# for better visualization
HTML\
("""
<style>
display: flex;
flex-direction: row;
flex-wrap: wrap;
max-width: normal;
white-space: normal;
</style>
""");

But it seems to not have pretty much effect... 😦

It is all messed up: https://www.kaggle.com/code/rubenvisionary/kaggle-5-day-gen-ai-day-4-g-search-grounding#Introduction 😦

[plucky harbor]

Could be a couple of things:

• Possibly a browser caching issue or a custom user setting; try opening in incognito mode, or a different browser.
• Possibly a system caching issue (unlikely, but possible); try a reboot.

[worthy pebble]

Hi all, have you ever needed an object detection dataset with specific objects/classes that just you just couldn't find? I made a tutorial were I explain how to make gerate such a dataset from scratch using controlnet, and I would love to hear some feedback; https://www.kaggle.com/code/koenbotermans/what-to-diffuse-free-coco-annotations

Have you guys every worked with synthetic data? Or generated anything that you have used to train object detection models for?

[opal shard]

Why are others unable to enable internet after copying my shared notebook?

[mint yoke]

Hey, i got the same project as Australian university guy, our company heavily invested in that guy for this ai project, and he handover the project to company and company handover it to me, after few observation is found out that this guy made garbage, lots of redundant same features in different form and results are not good. What should I tell company? Is this garbage bullshit or what language should I use to explain them

[zenith basin]

You should say it but obviously in a polite way. The most important is to explain your reasoning on why you think the project is not good the way it was built and the solution

[formal tendon]

"This is project is cooked like a well done steak"

[mint yoke]

Thanks

[mint yoke]

Thanks

[fast lynx]

Hai
i am currently developing a website as for the backend i have started with the porstgreSQL but now my client ask he want to store images so what could i do

could some one please help me

[nova merlin]

Can't u store images via postgresql 🤔

[void moon]

did u use claude for ur help?

[vale pine]

https://www.kaggle.com/code/bhaskarmishra44796/predict-podcast-listening-eda-ml

Can anyone look into my model and tell me what should I do to improve my model's performance?

[crude tide]

@restive frigate Hey sorry to disturb you but I had an issue regarding the request of a copy of my data from kaggle
I tried to contact the support but not no answer in return.

[brazen pond]

no

[nova merlin]

There r too many scammers in the server unfortunately we can't do anything

[rocky spoke]

Hey guys I’m working on the Isic dataset and need some help(using tensorflow btw)

[crisp void]

Has anyone received the badge for completing the Kaggle GenAI course? I just did, I don't know if I am on the right track.

[tropic current]

I didnt get one

My teammate got it

[neon aurora]

literally a ML community. Just train a model and implement it in a discord bot to ban them automatically. Fun side project 👍🏻

Did you guys see the github repo with all the system prompts to top AI models? https://github.com/x1xhlol/system-prompts-and-models-of-ai-tools - hot topic for the past 48h

[empty prawn]

Que tal gente!, les comparto mi kaggle link :: https://www.kaggle.com/gustavoaltamirano

[little heath]

hi guys we didnt create our notebook here in the competition, we created it separately but its public and we submitted the correct link. it will be reviewed right?

can kaggle staff/genai staff help?

[little heath]

yes, i got it

[silent nacelle]

How is this guy having 530 upvotes (250 non-novice) and still remain to bronze level?

Isn't there some kind of vote manipulation?

[rich halo]

tesfamichael9428 has been warned

Reason: Bad word usage

[waxen pendant]

From Vietnam, want to make a robot to assist physician in clinical problem.

[open aurora]

From Bangalore - Senior Automation test engineer.
Transitioning into AI Engineer or Data scientist

[stoic condor]

From Greece live in Luxembourg - Reliability Automation Engineer
Interested into transition into Data Scientist

[livid fossil]

hey

[drowsy granite]

Hi

[old glade]

Hello! I'm an USA electrical engineer focused on tech relevant to robotics.

[vale pine]

💀

What in the devil's Den is this

[modern kraken]

tffff

[empty arrow]

Hey guys, sorry for the stupid question, I'm doing some research about the best selling games of all times and facing some trouble finding the most adequate dataset. I'm sorry if i am in the wrong channel, but could someone please give me some directions? Thanks!

[vale pine]

Breaking models is my speciality

[signal crag]

[signal crag]

whats the problem in the data set you want to work on ?

like what exactly is causing an issue ?

[empty arrow]

I didn't find a good one yet, thats the problem, the only ones i found were in some unknown unit, and it was different from other sites that I used

[neon aurora]

Probably won't be the full dataset, but maybe try to look into Steam. You can probably check for number of downloads/game purchases

The issue is it will be probably PC based only

[signal crag]

hmm

can u share the dataset here

or in dms

i will try to check the problem

[mint yoke]

Anybody have great success in stock market using time series forecast?

[little heath]

are the winners out

[ionic peak]

Hi 👋
Anybody interested in AI/ML projects?
Ping me if u r interested

[zinc moth]

@ionic peak

[ionic peak]

Yes?

[regal totem]

hey everyone

I am new in ai field

I know some basics concepts like what types of learning, challenges and errors we face in machine learning

is that a good start?

[signal crag]

u can refer to a book called "hands on ML with scikit-learn , keras and tensorflow"

i cant send here cause i dont have nitro and its more than 50 mb

kaggle also has its own section for learning

andrew NG has courses on DV which can also be a great start

[clear lotus]

Awaiting for generative ai capstone project results @9:30 pm IST

GOOD LUCK EVERONE

@regal totem when you started we can learn

Learning is endless

[jagged fractal]

Notanymore

[ruby dagger]

hiii guys

[zinc moth]

Hn

[ionic peak]

Ping me

[wicked berry]

hey

[silent nacelle]

When you spam tqdm bars

[bright furnace]

Hi everyone! I am new to kaggle. If I want to build a profile in CV, where should I look? Any specific competition suggestions are welcome !

[verbal bramble]

Hi, kagglers.
Kaggle is one of the most knows contributors and pioneers in ML and Data field,
and I would like to point that ensure ethical and responsible AI and good data governance it is role everyone need to maintain from tech and non tech individual alike.
And for this reason I am inviting you to stand against misconducts that can bring harm.

Recently, I come across some people sharing restricted high risk content on huggingface for militay usage and applications.
Could you understand the underlying meaning of this.

Going silent about this , is unbelievable.

Immediate actions are required.
https://huggingface.co/datasets/ZennyKenny/tactical-military-reasoning-v.1.0

I urge kaggle community as an active community to report it.

[onyx salmon]

finally i can post here

[drowsy granite]

Hello Everyone، I want to increase my experience and need to do more complex projects, Are any one interested.

[lucid flame]

Newbie question: Do ARC Challenge submissions need to be in Python (or R) or can other programming languages (such as Common Lisp) be used?

[heady hedge]

Hi. I am Aman Dubey, from India. I have done Bachelor of Engineering with Computer Science. I now want to transition into Data Science.

[normal talon]

Hi there folks.

Here is Raf from Brazil.
Just a dev learning more about data.

[true tapir]

Hi Folks, Great to be a part of the community, looking forward to learn and contribute.

[slim torrent]

hey everyone

[ebon quest]

Anyone doing Spaceship Titanic challenge ?

[timid stirrup]

So i've no knowledge at all in coding, but i've used the differents llm to produce data analysis scripts since 2022.

The last script i've been working on is Mamba based, it is trained on BTC/USDT on the 15min dataframe since 2018.
Now, over the night i got this

Signal Accuracy
Approximately 70.8% of signals correctly predicted the subsequent price movement:

20 out of 29 UP/LONG signals were followed by price increases
8 out of 11 DOWN/SHORT signals were followed by price decreases
6 out of 9 NEUTRAL/HOLD signals correctly identified sideways movements

My algorithm was specifically good at loosing money, so i inverted the signal lmao

used mamba minimal also
https://github.com/johnma2006/mamba-minimal

[timid stirrup]

Example of the last signal at 95059 and now it is under 94740
and i think i've found a way to improve the signal accuracy again

[manic haven]

Hello everyone

I am kinda new to ai
Rn i know basic python, ML, NLP and currently learning DL

can someone guide me a little
I got some questions for this field

I am in my btech 2nd year

[bronze wind]

Hey Kaggle folks, I need some quick help. Could you recommend a structured and beginner-friendly generative AI course for someone already familiar with NLP and ML?

[bronze wind]

Start building projects, contributing to open-source, and participating in Kaggle competitions, it definitely adds a lot of value

[manic haven]

yeah i am participating in kaggle competetions but i dont have any team i am looking for one so i can learn from my team members too

[bronze wind]

Try asking in looking-for-a-team, you might find one.

[manic haven]

yepp doing that rn😅

[slim torrent]

i heard currently pandas and bs4 etc are outdated and replaced by more powerful libraries like polar, crawl4ai etc so anyone who knows or is a data analyst can you pls guide me on which libraries to learn as well as i am completely open to learn under someone

[wicked helm]

hey
started learning dl...
if someone interested to learn in a team..
we can follow up

...i am following aurelien geron.
the famous
hands on ml with scikit-learn kearas and tf

if someone is on same stage as me..
feel free to ping 🌹

[wicked helm]

umm
yeh these are more modern tools for data visualtions and webscraping...
u being such aware on a begiiner stage is good..
i would suggest you can go for it..

[slim torrent]

i am intrested

[wicked helm]

whre u from?

[slim torrent]

India

you?

[wicked helm]

whre in india

[slim torrent]

Rajasthan

[wicked helm]

oo nice,,

[slim torrent]

u?

[nova merlin]

U too Indian 💀

Jaipur ?

[polar cape]

Hi

I am a data engineer

Wanted to know the easiest way to learn Machine learning and Deep Learnign within 15 days

[slim torrent]

no

[nova merlin]

Kota ?

[slim torrent]

no, bro i aint gonna give you my excact location

[nova merlin]

Lol ok

[tardy osprey]

How to upload notebook on kaggle from GitHub?

[night shoal]

hello

[split siren]

hi

[heady bone]

Hi, am interested

[nova merlin]

Didn't u try any ai trading assistants ?

[craggy trail]

Can we learn together

[bronze kite]

hi eceryone, is there anyone from suport? i have a simple issue on my kaggle

[clear lotus]

Let me know the issue | I will try to resolve | I am not from kaggle support | Your fellow learner

[short wedge]

There is no kaggle support on discord, you need to email support or ask in the forums.

[mild scaffold]

Those who have uploaded their project on Kaggle, even if their name is not in the top 10 competitions and in the Honorable Mention, will they not get a certificate?

[quaint creek]

I have’s received the badge/certificate for the Google project. I tried to contact the email address mentioned but it is bouncing back

[rose violet]

hi, typically what does the interview process look like for machine learning and/or computer vision roles, for both internships and full time

[rugged plank]

Hey I just finished, Andrew Ng's first course on machine learning specialization and wanted to practice linear regression and after logistic regression is there any competion or dataset I should use?

[polar cape]

hey

Hi people

Looking forward to know How can I learn ML in 20 days being a Data Engineer

[hollow dock]

Well i just received a 170$ invoice from Google Cloud, i ve only used it for the the Gen AI course , did someone had the same issue ?

[nova merlin]

Where currently ?

[polar cape]

Means?

[nova merlin]

Data Engineer at ?

[polar cape]

A company

[nova merlin]

Yes bro I'm asking company name

[polar cape]

Wont disclose

[nova merlin]

If it's not quant then why not

Oh ok atleast u may tell why u wanna pivot from data engineer to ml engineer or data analyst or data scientist maybe

[polar cape]

I have curiosity to learn

[nova merlin]

And I've curiosity to learn/ask

[polar cape]

How is your learning goinn on

[nova merlin]

Moderate

[polar cape]

You reading or watching videos

Because videos take time

[nova merlin]

Books n articles

Yeah it hurted eyes n time to watch videos except the ones by Andrej karpathy

[polar cape]

Can you suggest some

[nova merlin]

Won't disclose

[polar cape]

Great

[wet beacon]

@radiant cloak hi

[near peak]

what's good guys

[rose violet]

hi, i'm looking for a dataset to train a license plate classifier. i want to first train a model to find the top left and bottom right coordinates of the optimal bounding box and then train a model to classify the characters of the liscense plate. are there any kaggle datasets any of you know of where the labels are specifically the top left and bottom right coordinates? most of the datasets i've found aren't clear in the description what the coordinates are

[south oracle]

hi

hi

[agile cave]

Sounds like well known MNIST problem, but for license plates

[rose violet]

kind of

except there's also the added problem of object localization

and then there's multiple digits/letters that need to be classified

[agile cave]

yeah, so sove that first, then apply MNIST solution to data you take in

[rose violet]

so unlike an artificial neural network won't work like with mnist

[rose violet]

the problem is i need data

i can't find a proper dataset

there are many but, none that i've found are clear with the format of the labels for coordinates

[agile cave]

So you have two problems: object detection and then classification

[rose violet]

yes

i think it's object localization -> object detection -> classification

first find the location of the license plate, then identify the locations of each individual letter/digit, and finally classify each one

[agile cave]

What are we talking here? A camera on a highway taking snapshot of cars?

[rose violet]

yes

[agile cave]

So first, you'd have to have the system choose which frame from video feedback to use or discard.

This is fiinding the frame that has a liscense plate in it

Its trivial after that

Then you can do your row by row analysis of the plate, in a manner of speaking

If you know the amount of digtis/ letters per plate, you don't need to ID each individual one. You have a NN that has multiple output layers for each in a series, and as it processes rows of the data (the pixels), it changes output weightings for each position, and final scores for each output layer gives you the letter or number for that position (intermediate outputs are ignored). Use LSTM or RNN layers in your network (so, as it processes row by row, it is "mulling" over a final answer). While training, perform error correction only on final answer, not the intermediate ones. Make sense? As to data, go out and take pics of stationary cars, some pics in which license plate is not in pic, others when it is in. Useful for first step I described (choosing video frame)

[rose violet]

oh i'll probably just have it set so users can input an image

but i might have it detect from a camera later on

[rose violet]

isn't that inefficient though

searching row by row with different sized bounding boxes

cause you'd have to go through each interval and then also try with different sized boxes

[rose violet]

yeah that sounds like a good project idea but is a bit more advanced than what i had planned

i have a lot to learn before that lol

and for the problem of getting the right frame, would using reinforcement learning be a good idea

i don't really know much about it but this seems like a reward based problem

[rose violet]

also do you think it would be a good idea to do this in c++

so that i can do it from scratch while still making it run efficiently, and it would help when i want to run this on a camera so that it can work with microcontrollers

[weary karma]

hi

[cobalt folio]

*📢 Seeking Research Internship Opportunity in Machine Learning (UG Student) *

Hi everyone! 👋

I'm currently an undergraduate student passionate about Machine Learning and AI. I’ve recently submitted my first research paper, which is under review and expected to be approved soon. I'm now actively looking for a remote research internship opportunity where I can contribute to impactful ML projects and continue learning under experienced researchers.

About Me:

💡 Strong foundation in ML, deep learning, and Python

📄 One research paper submitted (soon to be accepted)

🛠️ Hands-on experience with PyTorch, TensorFlow, and common ML libraries

🧠 Interested in areas like NLP, Computer Vision, and Generative Models

I’m eager to collaborate, learn, and contribute meaningfully. If you know of any open opportunities or labs that are looking for motivated undergrads, please feel free to reach out or connect.

Thanks in advance!

[lethal otter]

Idk if this is the right channel for that

how did u become a Microsoft student ambassador tho, and is has it been beneficial?

I’m starting college in the fall lol

[cobalt folio]

Yeah, just googled benefits of MLSA

[lethal otter]

How hard was it to get in tho

Ya gotta any tips

[slate cloud]

i wanna read this paper is there a link?

[mint yoke]

We have dataset normal that is after replacement of new idlers, we have abnormal (faulty) data before replacement of idlers. Data is in the form of real positive fft data that is it's in the form of list per row, also have DC, and peak to peak data too. We have long conveyor approx 5km long in that we have 1000s of idlers, we replace them when we see abnormal vibration and noise with the new one.

I tried unsupervised vae base anomaly detection trained it on normal data and tested it on abnormal data but it won't generalize. It works let's say I trained idler 1 and tested it on fault idler 1, but i want model that will generalize across the conveyor let's say we trained randomly 100s idler it should generalize on all idlers , all idlers have same dimensions. What else we can do ?

@tender bluff

@normal vessel

@grave karma

@tender bluff

@gentle iris

@solid python

Sampling rate 10khz

Filtered good signal based on DC and peak to peak values like above 4000 peak to peak considered good quality signals

And above 8000 DC values considered good quality signal

@shell haven

@magic condor

[radiant nymph]

Hi guys, im new in DATA SCIENCE and AI Field I just understand ML Models regression classification clustering using sklearn then i move to deep learning ann and cnn using keras what you guys suggest me how to successful in this field ?? any road map?

CV and NLP is also another field how to know which one is better for me and im also new on kaggle please guide me

[opal shard]

First rule: Don't ask for guidance. Be the guide.

[rose violet]

in my opinion, the way you know is by starting to learn both and then see which one you are more interested in

[opal shard]

The field of technology is different

[hard shale]

Hi everyone! I'm Prameela, currently transitioning into data science from a testing background. I've recently completed a machine learning certification and am exploring Kaggle projects—starting with Titanic Survival Prediction! Excited to collaborate and grow with this amazing community.

[sacred shale]

Hello everyone! I hope you’re all doing well. My name is Ali, and I’m new to the IT world. Right now, I’m focusing on completing the learning section on Kaggle, and then I plan to dive into real data science projects. Not sure what to expect, but I’m excited for the journey ahead!

[summer oracle]

Can I also be a part of this learning?

[round iron]

Hey everyone my name is Salmin, see you on leaderboard

[wicked berry]

@chrome rover may your check your DMs

please

[frail wigeon]

Hi. I am Nas. Looking forward to learning

[hidden depot]

Hello I am sam. I cant wait to learn these concepts and compete

[dense charm]

Can I also join? If possible

[frank cloak]

Hey guys its my first competition, and I want to make a submission. I disconnected the internet from my kaggle notebook settings and am trying to save a version of it. But its not working, i just click on the button and nothing happens. Also there's a red Failed to save draft written. I also tried submitting the notebook to the competition but it does not seem to be working. I just click the button and nothing happens. What should I do?

[west bone]

hi'

anyone here who is learning data analysis?

[tribal heron]

So the reason why alot of kaggle notebooks are bare skeleton is because there is a 1MB limit to submit/save the notebook 😦

[lilac yoke]

UPDATE: {RKaggle} Version 0.1.1 is now on CRAN.

RKaggle can now load .shp, .geojson and .feather files.
https://github.com/benyamindsmith/RKaggle

[stiff widget]

what's the best aproach for hosting torch models? i want to deploy my model and host it with an API but idk how...

[signal crag]

nas is my fave rapper

[signal crag]

i think we all are , thats why we are on this server ?

[quick brook]

I would drive anyone absolutely nuts if I learnt in a group 😂

PCA? Nah we need to know the entire math behind it first!

[neat pagoda]

Hi, here to learn from all of you, hope it will work 😀

[nova merlin]

-# keep on hoping

[gentle kiln]

Hey everyone

anyone here who is working as a business analyst

I recently started working as a business analyst, i'd like some inputs on how to improve critical & analytical thinking and business sense

[gaunt sonnet]

Hi Im newbie

[wicked berry]

Hey 👋

[hybrid meteor]

HI, I am here to learn from all of you

[fiery hearth]

Good day everyone

[pure berry]

Hi people, I have created a very beginner friendly notebook, following my venture in building Agentic AI workflows using CrewAI framework. Please have a read of the same and if you find it useful, please comment and upvote, so that the notebook reaches to a wider knowledge hungry audience.
https://www.kaggle.com/code/sudharshanravi/ollama-crewai-llm-agents-on-kaggle

[tiny torrent]

hello

[inner stag]

Hi I have a question for all of you.
I am a new data scientist, but I dont have any practical job experience yet.
Are Kaggle notebooks enough as a portfolio?
I am currently Kaggle Notebooks Expert should I start applying for jobs now?

[hard wadi]

I have a question regarding LLMs. Is there such thing as Prompt Grounding. Additional to a prompt (created via careful prompt engineering), I have also provided a csv file to help ground the LLM (so the LLM could go through the csv and choose its answer). If you guys know what I am talking about, please let me know!

[ivory tapir]

odd question to ask anyone here work as a consultant or analyst
need some pointers to gather deck making training material

any help would be much appreciated

[serene coral]

Hi

[left radish]

hey all, just wondering if there is a way to save the processing of a dataset for use in later sessions of a notebook?

[agile cave]

Yes, you can save data to Folders in your Kaggle account. I believe it starts with this: Go to “Data” tab (right sidebar in the notebook UI) → Click “Create Dataset”

[vast girder]

Women in AI is inviting applications for the Women in AI Awards! Are you (or someone you know) using AI to lead change across Asia Pacific? Apply today!
https://x.com/__Vaibhavi/status/1922338371159855227

Vaibhavi Singh (@__Vaibhavi) on X

@women_in_ai is inviting applications for the Women in AI Awards! Are you (or someone you know) using AI to lead change across Asia Pacific? Apply today!

Link in the comments! Pls amplify! @AnitaBorg_India @WomenWhoCode @wimlds @pyladies @WomenTechmakers @womenintech @UN_Women

X · May 13, 2025, 5:09 PM

[nova merlin]

No thnx it's no code one

[valid kayak]

Hey, I uploaded my first data set. Check it out 😋 https://www.kaggle.com/datasets/alshariefabdalla/synthetic-e-scooter-trips-1k-sample-2024 Feel free to 👍 + Download the dataset 🙂

[valid kayak]

🚨 Warning to Members Regarding Fraudulent Activity

Hey everyone — I want to make you aware of a serious issue.

A user in this server approached me privately and asked me to share my ID, LinkedIn, and email so they could impersonate me to apply for jobs on my behalf. They claimed it was a “golden opportunity,” but this is blatant identity fraud and illegal in most countries.

They said they “know how to bypass all kinds of risks,” which is a huge red flag.
If anyone else receives similar messages, I strongly advise you to not share any personal information and report the user to the mods or admins immediately.

Stay safe out there — if something feels off, trust your gut. @candid rapids @chrome rover

[gilded lava]

hi !

[chrome rover]

Thank you for bringing this to our attention. I've removed the user from the server - and I definitely think you made the right choice.

[valid kayak]

No Problem.

[wicked berry]

hello

[astral anchor]

HI, I AM ML ENGINEER FROM INDIA

[signal crag]

where did u do masters from ?

or did u finish btech in cs with ai/ml spec ?

[lofty hound]

Hello everyone, I'm excited to learn more and more about AI 😇

[vital charm]

hey everyone - quick question: does anyone know if algoverse ai is worth it?

[limpid river]

Hi

[nova merlin]

thnx

[rich hemlock]

what do you think?

[stuck pecan]

Hey

Who's participating in the new hackathon?

[nova merlin]

The ones who are actually participating they r near first progress prize already ...

[stuck pecan]

Well they are really good i guess

What about you?

[nova merlin]

Really bad found too complicated

[marsh kernel]

🚀 Looking for Teammates – Join AgriIntel! 🌱

Hi Everyone,

I’m putting together a high-impact AI project — AgriIntel — a Smart Farming Assistant built to solve real problems faced by small-scale farmers like my own family.

👉 This isn’t just a side project or a college assignment.
AgriIntel is being developed as a serious portfolio product to help each team member showcase real-world impact, demonstrate ML/Data Science skills, and land strong remote roles.

---

🧠 What We're Building:

• 🌾 Crop Recommendation Engine — based on soil, pH, rainfall, etc.
• 📈 Crop Price Forecasting — using Time Series (Prophet/LSTM)
• 🍂 Leaf Disease Detection — with Computer Vision (YOLOv5 / MobileNet)
• 🗣️ Hindi Voice Assistant — powered by Whisper + gTTS
• 📊 Insightful Dashboard — Streamlit or React

---

✅ Why You Should Join:

• 📌 Build something recruiters will ask about
• 💼 Boost your GitHub + Resume with real-world work
• 🧩 Collaborate in a sprint-style, outcome-driven team
• 🌍 Contribute to a product that impacts real lives

---

🔗 Full Pitch + Roadmap:
👉 Click to View

📩 Interested in joining? DM me directly or connect on LinkedIn:
💼 LinkedIn – Dinesh Kumar

Let’s build AgriIntel together — and create something that truly matters! 🌾💡

—
Dinesh Kumar

[frank cloak]

I've actually kind of thought about this for a while. And I agree, but let's see if our reasoning matches

[signal crag]

i mean if you're using synthetic data for a model u r creating from scratch it can be great

i use that data to work on my logistic reg model i created from scratch

[stiff widget]

i got a job offer to make a model that answers call center calls, and honestly i've never been more lost. any tips?

i've mostly worked with n-grams models and simple nlp stuff but this sounded a tad bit big

[kindred lichen]

All, Have an interesting job opportunity with a well known product vendor for networking product support. Experience looked at 5 to 10 years in networking, routing, switching, protocols, fundamentals. Please reach out to me at [e-mail]

[agile cave]

This book may be useful for crafting code that harnesses existing LLM's for a/your purpose:

... and if you want to learn how to build an LLM from scratch, check this out:

[fleet cobalt]

https://www.linkedin.com/posts/nikhil-gupta-62b024210_nlp-bert-transformers-activity-7330126410872696832-sw6G?utm_source=share&utm_medium=member_android&rcm=ACoAADVlgIMBcI5X9Utxf6tAt8MKeT6IzX9YSVY

Pls like, comment and post

[tawny marlin]

Hi, i wanted to ask about Kaggle courses, I couldn’t find any license under it (courses like intro to machine learning) am i allowed to use code and learnings from it commercially and for non-personal use? I hope an admin sees this and replies

[hardy trench]

hello , my name is Samuel

[wide ridge]

Hello, I’m new here 😇

[undone cape]

hii

[daring sonnet]

Hello. My name is Saad and im an aspiring data scientist

[signal crag]

fair

[signal crag]

bro's really linkedinfarming here 😭 🙏

[hybrid shard]

Hi, my name is Faisal, Tech project manager passionate about AI and cybersecurity. I believe knowledge is meant to be shared, not stored.

[frank cloak]

You're absolutely correct faisal

[teal slate]

Hello

[olive ember]

Hello anyone applying for interships? If yes please let me know

[fickle vine]

hey, does anyone have info regarding amazon ml summer school 2025? when will registrations start, dates or any other info? i searched but couldnt find anything anywhere

[hallow whale]

Yes trying to get one 🙂

[wicked helm]

https://docs.google.com/forms/d/e/1FAIpQLSf-oaubYrc8SkKUxRuGg9oKWsvW42bI8_j9IcRvh4k1fJpjLg/viewform?usp=dialog

i am working on a major project..
i need your suggestion for proressing further..

it will hardly take 10 seconds..

[plush coral]

hiiii

how do i start i only know basic python

[agile cave]

Plenty to work through on Kaggle site, and get you sorted out: https://www.kaggle.com/learn

books you may want to check out too, relevant to AI/ML:

@plush coral Welcome to the Adventure, btw

[plush coral]

oko thanks for guidance

[lyric widget]

Can anyone suggest me what should be done first in machine learning ? Mathematics or practicle implementation .

[wide ridge]

I recommend learning the basics of mathematics, such as linear algebra, probability, etc

[lyric widget]

Thanks @wide ridge

[agile cave]

Start with this book. Will cover all algorithms you will deal with, in the form of easily digestible Diagrams. Should be the Bible of AI/ML to anyone coming into the field, imho.

from it, you'll know where to go next/ what to explore more deeply

[lyric widget]

Thanks for sharing Sir !

[agile cave]

Sure. No problem. As a side note, master Python and it's AI related libraries, which is what most use in the field. For this, might I suggest these. With all three books I have mentioned, you can establish a solid foundation/ grasp of it all.

Good luck

[mortal rain]

Hello! For anyone interested in fantasy football or NFL sports betting, I'd like to share the dataset I've been working on for the last couple of months. There are over 1000 features of weekly and yearly player and team statistics that make it great for predictive models or general data analytics, so if you're interested in sports statistics or like to dive deep into robust datasets, then I'd highly recommend checking this one out! Any feedback would be greatly appreciated. Enjoy!

https://www.kaggle.com/datasets/philiphyde1/nfl-stats-1999-2022/data

[onyx lynx]

has anyone worked with shopify development using RemixJS and Prisma with a postgres backend? please dm Im looking for some help.

[native briar]

Hey can anyone give me some tips for an interview that I’m gonna give next week, job’s about Data Science, idk if they’ll hire me to do EDA or ML etc etc… should I specify or smth?

[wicked berry]

ping

[austere briar]

TradingView Premium Full Free version (Windows + MacOs): https://www.reddit.com/r/TVFreeHub/comments/1knfko8/

[cosmic lagoon]

https://medium.com/@satvikg21111124/smart-tasks-smarter-agents-9509aa85f4f0

[sudden furnace]

Hi there, Myself Saad Web scraping learner and practioner ...

[lime seal]

Hello

[ivory tapir]

How's deeplearning.ai courses

Heard they go in depth in mathematics as well as feature selection

[ancient shell]

is it possible to turn on code suggestions on kaggle notebooks?

or atleast function annotations (so e.g. when typing sns.pairplot() i'll see what arguments it takes)

[bitter bobcat]

Hey, new member

[marble elk]

Hello I am new to this field and I have done traditional ml and little bit of dl so I want to know what path should I follow going forward in this journey?
Also I am preparing for placement season so is there any way where I can make project and gain certificate for completion of the project.Thank you

[chrome raptor]

a noobie bioinformatician, but an ambitious one

[ruby dagger]

Hey everyone!
I just started a new discussion on Kaggle and would love your thoughts and feedback. Feel free to check it out and join the conversation! 😊
Discussion link (https://www.kaggle.com/discussions/general/580462)

[little sage]

This is being passed around. Keep it on the DL. Lawyers, news outlets, and law enforcement are combing over this report and the data.

Take your time to digest it.

Reader discretion advised.

Is Tesla's FSD missing the road for the trees?

https://www.alshival.com/is-teslas-fsd-missing-the-road-for-the-trees

[deft hare]

Hey I am new to ML. I am from non technical background but ML has my interest.

[waxen solstice]

hi everyone - looking forward to conversations here. I'm based in SF Bay Area and have cybersecurity as background, new to ML.

[fervent hawk]

Hello everyone, I want suggestion for setting up gpu desktop for DL model training. Please can anyone suggest what configuration shall I go for? Consider a medium budget

[nova merlin]

https://blogs.nvidia.com/blog/jetson-generative-ai-supercomputer/

[wicked berry]

RTX 4070 + Ryzen 7 + 32GB RAM + 1TB SSD = solid medium-budget DL setup.

[cinder python]

Hi guys, I have 2 project ideas and the fun fact is both are completely different from each other. I'm confused about which one to work on. Can someone guide me on which idea I should choose? It would be great if someone from India could help, as my English isn't that good😅.

[glossy bloom]

any one knows why kernel is dying every time ?

im fine tuning YOLOv11s

GPU P100

[broken marsh]

Hii

Guys

[nova merlin]

@grok it

[umbral scarab]

Hi @everyone
I’m a first-time contributor.
I want your help to understand how I can contribute to an open-source project.

I am mainly focused on Machine learning and AI, I work in Python and related libs.
I want to understand how something works. I’d appreciate any help and would love to get more involved.

[umbral basin]

helllo everyone

is there any cloud enthusiast

[halcyon solar]

Bop it 😏

[still isle]

Good mornig everyone

[elder wolf]

Hello guys. Is there anyone here who needs help with anything digital/ tech related? I need to raise 46$ for my rent before I get evicted in a few weeks. I also need money for food as stock is almost finished. Please help me as I don't have any friends or relatives, just relying on kind online strangers

[halcyon obsidian]

Hey Everyone
I have experience with DevOps, but I want to learn more about MLOps. Where can I find relevant blogs, MLOps certifications, and official MLOps pages?

[snow charm]

is RTX 3050 4GB + Ryzen 5 5600H decent for ML and DL ? (got this laptop 2 yrs ago still works fine)

[unreal trout]

Hello everyone, I am looking for a technical co-founder if anyone is interested in building a startup. Hit my DM will explain the idea there.

[nova merlin]

https://medium.com/@NirZicherman/stop-looking-for-a-technical-cofounder-c1cd76a29854

[halcyon solar]

I am the technical founder 🤯

[obtuse creek]

https://x.com/alve_om/status/1927669973402472850

Om Alve (@alve_om)

💸 Puch AI is PAYING ₹500 to 5 lucky winners!

No catch. Just try out Puch — Made for Bharat 🇮🇳 Fluent in every Indian tongue

Link in comments.👇

X · May 28, 2025, 10:15 AM

[ruby dagger]

Hey everyone!
Checkout this discussion on Kaggle and would love your thoughts and feedback. Feel free to check it out and join the conversation!
Discussion link (https://www.kaggle.com/discussions/general/580462)

[dense kindle]

Hi guys I just started with kaggle , I was looking at titanic problem

[west plaza]

Hello, I'm new in Kaggle and I would like some tips that would help get started

[frank cloak]

First kaggle competition, very disappointing experience. Couldn't even make a submission based on my prediction

[cosmic prism]

How?

[valid flame]

hey everyone , i am new to kaggle glad to be part of this server

[little sage]

Fighting evil by moonlight.
Writing code by daylight.

TechnoKing, come and taste the fists of the King born from the woman dressed with the sun, with the moon at her feet, and twelve stars in her crown.

https://www.kaggle.com/code/alshival/is-tesla-s-fsd-missing-the-road-for-the-trees

[hybrid axle]

I am looking for a co-founder who resides in New Jersey, Please DM me if anybody interested, Thank you

[nova merlin]

Where do u reside

[hybrid axle]

NJ

You ?

[nova merlin]

No outside states

[pale sigil]

Hii

[stark kindle]

Hi

[heavy bay]

Hi everyone ,

i recently joined this server
nice to meet you all

[halcyon solar]

Welcome 👍

[wheat marlin]

Hii

[turbid abyss]

Hola!! 👋

[quasi cloak]

Hello everyone, I am dodge!

[true reef]

Hello Everyone
Nice to meet you all and hope we learn something together😃 .

[broken oracle]

Hey all — quick update on Rensa, a MinHash library I’ve been building in Rust with Python bindings. It’s focused on speed and works well for deduplicating large text datasets — especially stuff like LLM fine-tuning where near duplicates are a problem.

Originally, I built a custom algorithm called RMinHash because existing tools (like datasketch) were way too slow for my use cases. RMinHash is a fast, simple alternative to classic MinHash and gave me much better performance on big datasets.

Since I last posted, I’ve added:

CMinHash – full implementation based on the paper (“C-MinHash: reducing K permutations to two”). It’s highly optimized, uses batching + vectorization.

OptDensMinHash – handles densification for sparse data, fills in missing values in a principled way.

I ran benchmarks on a 100K-row dataset (gretelai/synthetic_text_to_sql) with 256 permutations:

CMinHash: 5.47s

RMinHash: 5.58s

OptDensMinHash: 12.36s

datasketch: 92.45s

So yeah, still ~10-17x faster than datasketch, depending on variant.

Accuracy-wise, all Rensa variants produce very similar (sometimes identical) results to datasketch in terms of deduplicated examples.

It’s a side project I built out of necessity and I'd love to get some feedback from the community 🙂
The Python API is simple and should feel familiar if you’ve used datasketch before.

GitHub: https://github.com/beowolx/rensa

Thanks!

[foggy hill]

https://x.com/hussamkazim77/status/1928865203556417819

Hussam Muhammad Kazim (@hussamkazim77)

Model: VEO 3

Prompt:

The camera hugs the ground, shadowing a focused ant dragging an oversized crumb. It threads its way through towering blades of grass, crosses a wide expanse of concrete, and slips into a thin crack along the house’s foundation.

#Veo3 #googleflow #aivideo

X · May 31, 2025, 5:24 PM

[agile cave]

You have reached the message cap. You have reached the conversation limit. 🤬...

[little sage]

This is a behavior I witness time and time again when training machine-learning models, and I am wondering if this community has an answer.

Last year, I trained an ML model to predict surgery volumes for a large medical institute. The model performed terribly at the surgeon level, but when I aggregated the results, they were so close. Less than a waiting room of people. I'd predict 950 surgeries and there would be 940. The aggregate level was good enough because there was an IV shortage, and so the hospital was able to manage their supply using the model. They didn't need the surgeon level predictions.

Fast forward to today, and I have another client with a different model, and the same behavior is there. Low accuracy at the nurse level, but we get really close at the aggregate level.

What are you thoughts on this behavior? This model is much more complicated than the previous one. This one uses nurse profile and job data, as well as conversational data (semantic embeddings).

[little sage]

Right now, I have a bunch of candidates in a batch. I wonder if... each candidate in its own batch would improve the model at the candidate-level. Idk. Thank goodness for Vibe coding because I'm feeling lazy. I'll just do it tomorrow. Monster Hunter Wilds is calling me.

[fickle ether]

xhttps://pk.indeed.com/viewjob?from=appsharedroid&jk=921ae9db7b094909

xhttps://pk.indeed.com/viewjob?from=appsharedroid&jk=921ae9db7b094909?

[dark flint]

I am trying to build a simple Streamlit app for question answering on a given PDF. What approach should I use? Which open-source LLM would be suitable? What model should I use for embeddings, and should I implement RAG or another method?

[subtle magnet]

Harold Kimmey - https://www.kaggle.com/kimmeyh - https://www.linkedin.com/in/haroldkimmey/

[subtle magnet]

Check out https://github.com/NirDiamant and examples: (streamlit example) https://github.com/NirDiamant/Controllable-RAG-Agent also https://github.com/NirDiamant/GenAI_Agents and https://github.com/NirDiamant/RAG_Techniques

[tight vortex]

My name is Sophie and I'm learning Machine Learning Engineering through the MLE specialization course on Coursera, taught by Andrew Ng. I hope to accomplish a career in the field, related to data science and creating models/algorithms that help people.

[inner notch]

Hi!

[little sage]

Waiting for a model to train. Before I found out you can rent GPUs, I had to wait because my GPU was training. But not anymore. My GPU is now free. Free!!! I'M FREE!! 😂

[steep sparrow]

Hi everyone!
I’m working on a gait type classification project using IMU sensor signals. I'm trying to classify 4 different walking types (e.g., normal, foot drop, steppage, unsteady). I used data from 32 participants – 22 for training and 10 for testing.

However, I’m getting a low F1 score, and I'm not sure how to improve it.

Does anyone have suggestions on how to boost F1 score in this kind of classification task?
Should I try:

Different window sizes?

Feature engineering?

Balancing the dataset?

Trying a different model?

Would love to hear your experience or tips.

[spiral kettle]

Heyy Everyone!
I'm Devansh and I'm an Data Science and AI Enthusiast.

I am pretty curious about the whole AI thingy so I wanted to try learning it from basics. I'm working on basic Data Science and ML for now.

Looking forward to working with you guys!

[little sage]

This is probably the biggest model I have ever built. It's for a national medical organization. Their dataset is about 200 gigabytes and the model takes in profile, placement, and conversational data. I added a transformer within the feed-forward for the semantic embeddings. I just call it feed-forward loosely. Abuse of language.

Yesterday, I was able to accurately predict aggregate (this many candidates out of that many will do X), but I couldn't get it to predict accurately at the candidate level (this particular candidate's risk is X). So I rebatched it.

[rich halo]

al_shival has been warned

Reason: Bad word usage

[little sage]

Bad word? Lol. What was bad about what I said? I just said I want to challenge the TechnoKing to a kaggle competition. He can use his super computer and I will use a steamdeck. Prove himself, or I will always see him as a neurosurgeon who has never done neurosurgery.

[stable sigil]

dont use seggs words here, the bot doesnt like it, other than that youll be fine lol

[elder wolf]

Hello guys. I’m in need of help, my life is almost coming to a standstill. I need money to buy food till next week when I’ll be getting money from my contract job. I will pay back if necessary, please help me anyone out there. I’ve exhausted all options physically and have nobody to help me that’s why I’m asking for help online as a last resort. My food stock finished on Sunday so I haven’t even eaten anything for 2 days now. If anyone can help me I would greatly appreciate. Any amount will help. I promise I will pay back if necessary

[queen phoenix]

Are you able to go to a local food pantry in the area you are located?

[elder wolf]

I’m from Nairobi Kenya so there’s no such here

[spiral kettle]

If international transfer was allowed to me so I would've done it. I'm sorry man, I'm unable to do it.

[elder wolf]

What about PayPal?

[spiral kettle]

I don't use it here in India.

[elder wolf]

Or crypto

[elder wolf]

Oh okay

[spiral kettle]

Nopee. I'm still 19 and I don't have enough income to buy any financial instruments.

[elder wolf]

It’s okay. Thanks for the concern though, I understand

[spiral kettle]

You're welcome.

[barren gyro]

Does anyone have recommendations on where I can find public datasets that I can use to show off my SQL skills for my portfolio?

[little sage]

We don't let teenagers use a calculator on parts of the SAT, so don't let your teenager use Tesla's FSD.

When a king overstepped, America threw tea into the sea.

Now the TechnoKing has overstepped, and we are burning his cars because doing so may save a life.

https://www.alshival.com/is-teslas-fsd-missing-the-road-for-the-trees

[halcyon maple]

Hi I am Ankush Jha, from India, I joined kaggle today only, will proceed with 5 days generative ai course

[sour spade]

Hi! Everyone! Is anybody hiring?

[nova merlin]

guys my friend needed an help with a survey kinda from students in states
https://buildpad.io/research/63YT5Zt
(kindly share with ur friends studying in states)

[thorn blaze]

Hello!

[agile cave]

hunh?

[foggy hill]

?

[foggy hill]

1500 ChatGPT Prompts for Free

[inner stag]

Hi, I am a Data Scientist and Machine Learning Engineer. I have worked on many projects on Kaggle and am now a Kaggle Notebooks Expert. I am looking to work on real world projects. is anyone open to collaborating or can help me get started?

[agile cave]

Interesting data set ✍️

[foggy hill]

I am thrilled to share that I have completed first Course of IBM Data Science Professional Certificate

[little sage]

I, Alshival, hereby apologize for picking fights with Elon Musk. I need therapy after investing FSD for years. That's all I am going to say.

I wish to be a peacemaker.

https://www.kaggle.com/code/alshival/is-tesla-s-fsd-missing-the-road-for-the-trees

[muted trail]

I just hate when I come across papers that suggest using FasterRCNN and use another model for segmentation within the predicted bounding box... frcnn's precision is sometimes so ass

[light dune]

Would anyone like to critique my resume?

I'm a data science undergrad, currently wrapping up an MLE internship. I've put together a resume in hopes of securing a second internship this coming fall or next summer, would appreciate any feedback.

Please shoot me a DM if you're able to help.

[nova merlin]

https://buildpad.io/research/63YT5Zt

🗞️ Hey! I’m working on a project called Commit, designed to help high schoolers get the best of AI to finish their college applications with ease. We’re in the early stages and need your feedback to make it better. It’s a quick 10-second survey—promise!
🎯This is a youth-led initiative, and by sharing your thoughts, you’ll help shape a tool that could boost your own college app. Students from any country are welcome, and your input will make a big difference.

[agile ferry]

Damn u r cookin somethin!!

[nova merlin]

Do share with ur abroad friends n juniors in schools/colleges

[agile cave]

What are we talking about? The College Essay? It should be framed as "A Hero's Journey", drawn from the prospective applicant's Life Journey, to date . The AI system should facilitate crafting such an Essay, through queries, etc., imho. Also could make a killer, AI powered, personal journal, that responds to user inputs, of an idea. We're all always applying to somewhere, in a sense.

[sudden furnace]

Hi Guys
I have develop a web scraper using Python bs4 and pyautogui which extract the tweets from X.com without any API
Here is the video link on my Youtube channel https://youtu.be/KIAprXC3i5g

I want you to see this video and mention your opinion about this script, I scraped more than 100 tweets with scrolling limit of 25 ... These 100 tweets are latest not a high volume but you can managed to scrap huge amount of data from X

[wicked helm]

hello hereones!
i have recently made an educational latex paper on "building neural network from scratch"

kinda tell something ..

[ruby dagger]

Heyy 👋 guys check this out new Discussion
https://www.kaggle.com/discussions/general/583178

[wicked berry]

Hey guys, I am in highschool and I want to go in AI, I was thinking if some guys could help me out with filtering what all things I can do now which won't be covered in my CS degree. I just know basic math nothing much

[wicked berry]

Hey! Is there anyone participating metal kaggle hackathon competition?

[signal crag]

woah @cyan crystal here

[cyan crystal]

?

[hot tangle]

Hey everyone, I'm joining from Kenya. Currently on the GenAI Course by Google x Kaggle 😄

[midnight nest]

hello, I am newly here

[nova merlin]

meta ?

[brittle wigeon]

Hey guys, can anyone please answer my question in the meta-kaggle-hackathon channel. I do have more doubts as well. if anyone from staff could answer my doubts, it would be helpful

[digital finch]

hello

[brittle wigeon]

Hey. What's an expert?

[tropic plinth]

hi

[merry latch]

Hello, I'm Raoul Mushitu. He/Him

[supple imp]

Found good as a teaching material. Went through, well structured. Appreciate your efforts and for sharing the doc too.

[digital finch]

kaggle expert?

[indigo dock]

I tried everything I could to debug it. Is there anything else I can do? Any help would be greatly appreciated.

[broken jay]

@restive frigate (sorry for the ping but) Does someone know who Im supposed to ping when I see a scam?

[silk flint]

hello! can i get help and suggestions regarding problems i am facing in my kaggle work?

[manic swift]

are the mods sleeping ?

[long mural]

Probably, I think

But, the thing is I think those scammers are playing a wild game.

It's not the first time I see those pics, i've seen them in other servers.

Oh... they left!

[quick brook]

FYI it's in every channel

[paper stump]

Hello, everyone.
I have a small question. I am getting this message on Kaggle's website:
We are experiencing site-wide availability issues
Is there a website that shows Kaggle's current availability status?

[rugged arrow]

same here , there is an issue

[paper stump]

I think the issue is fixed now.

[white harbor]

Hi everyone!

I help early-career developers land jobs in competitive markets by providing tailored interview coaching and job search tools. If you’re interested in boosting your hiring chances, I’d love to connect and share more.

[little sage]

I keep getting bans for sharing this. Read the analysis. Just because you try to hide it doesn't mean it will go away. It spread like wildfire here in Texas.

https://www.alshival.com/is-teslas-fsd-missing-the-road-for-the-trees

[elfin tree]

Hello, I am completely new to the data science/ML community, but I have a passion to learn and thrive within it! so any suggestions on how to start will definitely be greatly appreciated. Thank you in advance.

[wicked berry]

Bruh I think Value Error Input contains NaN infinity or a value too large for dtype('float64')

Replace this line,
empty_cols = [cols for cols in X.columns if X[cols].isna().all()]

With this
empty_cols = [cols for cols in X.columns if X[cols].isna().any()]

<<<

[indigo dock]

Okay, will surely try it. Thanks dude 💪🏻

[indigo dock]

I recommend beginning with Kaggle Learn resources, as they have been very beneficial for me.

[wicked helm]

thanks subramanian sir

[wicked helm]

anyone??
i want to work on project
i am searching for friends. or team.

i have decent knowledge of ml and backend.

i am passionate for ml and dl.

tell me which project u are wokin on

[indigo dock]

Finally debugged it. The issue was with the Id column.

[kindred prism]

whats the appropriate channel to post about projects/tools/packages?

[quick brook]

So within my personal frameworks, I'm wanting to begin operations by using Deep Feature Synthesis via Featuretools. I realize that for Featuretools to perform at its best, the data should be at a minimum, structured in 3NF.

I'm wondering if there's a way to automate some of the conversion of tabular data into 3NF, by explicitly declaring column variables. I'm not sure if there's a way to make a program that can determine one to one, one to many and many to many relationships. Have any of you explored this or have a particular process for dealing with this?

Ah, I see, Featuretools makes reference to autonormalize, I'll have to learn its limitations and give it a go, that's cool 😮 saves me a lot of time xD

[sullen whale]

Hello kagglers

[timber depot]

https://www.kaggle.com/code/xixama/amesa-house-predictor
Guys, please do drop suggestions !!

[wicked berry]

Oh sound's good ur done with it

[indigo dock]

Ya ^^

[wicked berry]

Best of luck for incoming milestonez

[echo mica]

hello!
can anyone tell me how to hide certain competitions from showing up on my public dashboard? I can't seem to figure it out.

[fast loom]

Hi Guys , is there some good programmers ? Let’s make a team , I have an idea we can work on it together. Backend developers would be even great

[nova merlin]

u on twitter?

[wicked helm]

yes man

just started few dayz back

for?

[nova merlin]

u r into tech only if u rin twitter

also u r real only if u 've atleast 0.1k followers

[wicked helm]

ya good place for tech enthusiasts

[wicked helm]

exactly this sucks..
i mean to gain followers..

how can someone post..
50 s of posts daily..

[nova merlin]

ppl getting funded in days there all kinds of influential propagandas running there

[wicked helm]

they just post anythingl.

[wicked helm]

ya at big level..

kuch bhindi.ai toh kuch pucho.ai
sb wohi se source ho rhe..xd

[nova merlin]

no dude u've not been long on the platform or dont follow good ppl otherwise u might ve known

each guy 've their own algo to post

[wicked helm]

tell me na..
i dont have must idea..
suddenly i got a dream..
for networking..

so i joined x

[nova merlin]

bhindi ai 🔥 hai ig

[wicked helm]

there i see.. people are ceo at 19/

how long u been on x

[nova merlin]

16

[wicked helm]

14

[quasi python]

Wherever I go... I see teenager Indians nowadays lmao.

[broken jay]

real

[quasi python]

Good to see teens getting into tech early tho.

[dire plover]

hi im new

[digital finch]

That is me 😭

[halcyon cairn]

Hey everyone I am new to this discord and want to excel in ai ml and data science

Hoping to find buddies on the way

[prime umbra]

hii new

[final herald]

Hey, I am Stephen. I am passionate about ML/DL. I would admire to connect with likeminds.

[quasi python]

It's really nice thing to be in such communities at yound age. Keep it up.

[wicked berry]

https://www.kaggle.com/code/syburg/heart-disease-prediction-analysis

[fathom gazelle]

https://summer.hack.club/bhc I want project ideas for this

[agile cave]

An AI that does the dishes and cleans the house??? By that, I mean focus on pragmatic things. Plants crops, or, builds buildings, are suitable candidates. What other practical applications of AI can you envision? Love if you'd share your thougghts on that. Anyway, good luck, regardless.

[tidal lark]

Thoughts on Geoffrey Hinton's interviews he's been doing for a while on trying to warn humanity of the dangers of AI?

[agile ferry]

foook kids!!

[fathom imp]

Hi

[kindred prism]

I’d appreciate if you guys check this repo out and star it. Thanks a ton! Also consider trying the package out if you have a gpu :p

https://github.com/Shaurya-Sethi/transqlate

[wicked berry]

How do you guys run your codebases with kaggle resources instead of just putting it all in one notebook?

[quasi python]

Sup!

[rigid niche]

I'll take a look when I get home and see if I find anything valuable for my research. Thanks for sharing.

[kindred prism]

Sure!

[kindred prism]

If you’re doing research on something similar I think I could help by pointing out some gaps in the current state of NL2SQL

Lemme know

[mystic tusk]

Hello :D

[rigid niche]

let me know if you'd like a more casual breakdown

but I figure you can feed this to your agentic system to extract value

[kindred prism]

Nice! If you’re looking to build your own system - I’d like to point out that I did not implement PICARD, something one can easily do because it’s an inference time technique- but adds latency so I chose not to use it.

Secondly, the fine tuned models today (mine included) are mostly fine tuned on SQLite related databases so they’re unable to capture variations in dialects- you can attempt to solve this by gathering and structuring data that is of various SQL flavours including postgres, oracle, etc.

I have implemented a dialect conversion process instead where after the model generates the SQLite query, I used rule based methods to transform to be suitable for other databases

[rigid niche]

Hey Shaurya, really appreciate the detailed response. That makes a lot of sense regarding PICARD — you're right, it's elegant but the inference latency tradeoff is real, especially for production or edge deployments. Your choice to handle dialect variation via rule-based conversion post-generation is clever — pragmatic and avoids retraining across fragmented SQL standards.

As for me, I’m actually working on a broader symbolic-intelligence system called the Salgado Information Matrix (SIM):
🔗 github.com/Salgado-Andres/Salgado-Information-Matrix

SIM isn't limited to NL2SQL — it's a contradiction-resolution framework that fuses G₂ symmetry, φ⁰ emergence, and Recursive Emergence (RE) to map symbolic collapse events across domains (archaeology, AI planning, knowledge extraction, etc.). But under the hood, a lot of what you're doing in TransQLate aligns with our kernel architecture: fine-tuned dimensional compilers, retrieval-augmented latent space navigation, and projection onto structured ontologies (in your case, SQL schemas).

Specifically, my stack includes:

φ⁰ Kernel Compiler: An 8D contradiction-aware projection model that maps observed patterns into symbolic action spaces.

RE-Agents (Recursive Emergence): Each agent recursively collapses contradictions based on local coherence, analogous to a schema-aware reasoning engine.

G₂ Pattern Tensors: These are used for feature alignment — similar to how you're dealing with dialect normalization across SQL variations.

We’re already using it for Amazonian site detection, windmill recognition from LIDAR, and a philosophical recursion engine (via Omega Theory), but we’ve been planning a text-to-action query language, which overlaps with the semantic planning you’re doing in TransQLate.

lets connect and we can further discuss 🙂

I'd start here:
https://github.com/Salgado-Andres/Salgado-Information-Matrix/blob/main/SIM_Axiomatic_Framework.md

[kindred prism]

I’ll definitely check it out

[kindred prism]

You can dm me if you’d like - I don’t have a LinkedIn

[rigid niche]

perfect, I sent discord invite we can have more slack collaboration, lmk if you have any feedback after you check it out. I know it's very abstract math, but octionions can be explained simply once you are exposed to it. Stay curious!

[sly berry]

Hello Kaggle Community...

Just a follow-up on an application we made for a chess competition.

We submit via this guide:
https://www.kaggle.com/competitions-research-grants?utm_source=reg&utm_medium=email&utm_campaign=competitions-research-grants-program&utm_content

[stuck pecan]

Guys anyone know about the Adobe hiring challenge?

[quick brook]

Seems like a wonderful project, going to stay subscribed to this.

Something I've often thought about, is if in the future there's a way to quantify how much from learned patterns, and noise is likely attributable to emergence.

Intuitively, I think that if this were measurable, it would be relative to the features themselves. Perhaps some kind of propensity to conform rating on features for how likely they are to align with a normal distribution given a change in samples. Just a 5 minute, 4am thought 😂

[rigid niche]

That 4am thought is actually fire 🔥 — and not far off from what we’re exploring. In SIM, we think of emergence as the residue of contradiction collapse — not just an output, but a signature of how features resisted or aligned under pressure. So your idea of a “propensity to conform” is spot-on — in our terms, it’s φ⁰ resonance vs. torsional leakage.

We’ve been experimenting with G₂ torsion metrics and entropy descent gradients to estimate how much of a system’s final structure is due to compression (signal) vs noise accommodation. The real challenge, like you said, is making that relative to feature domains — since not all noise is created equal, especially across semantic or symbolic layers.

Would love to bounce more on this if you're ever prototyping — your instinct is pointing toward a metric for emergent alignment fidelity, which might even be visualizable across kernel projections. 👁️‍🗨️

And yes — those 4am ideas? Often the ones closest to the source 🔌

[quick brook]

Yeah I'm one of those ideas kind of people. I could learn something new, I'll understand it very well, but I'll immediately forget the name of it. Names, directions, metadata, I'm so darn forgetful, I place little importance on it (not consciously). But because when I learn something, I try to learn it based on what I already know, sometimes I think of wacky ideas like that.

I think there was another theorem that suggested the lack of conformity is a better indicator of possible emergence.

On that topic it does pique my interest, I considered that when there's no interactions and it's something systematic or absolute, then a normal distribution becomes less relevant. If it's a system relative to interactions (such as a flight reporting system), you'd generally expect normal distributions.

But if an interaction-driven dataset with more data, doesn't move towards that normal distribution, it could suggest there's other parts that it could join to form that normal distribution with more samples.

For now I'm focused on setting up my framework to be as efficient as I can get it, because EBMs are not GPU accelerated 😅 but they're so powerful I still want to use them

[rigid niche]

That makes total sense — I get the vibe completely. Honestly, the fact that you prioritize conceptual structure over labels or metadata just shows you’re tuned into the deeper patterns. I’m the same way — I rarely remember the name of something, but if I can derive or reconstruct it from first principles, it sticks. That way of thinking might seem chaotic from the outside, but it’s often where the wacky (read: genius) insights emerge.

That idea about lack of conformity as an indicator of emergence is gold. You're onto something subtle — especially in interaction-driven systems. When more data doesn't push the distribution toward normality, it's almost like the system is hinting there's another layer — a latent interaction or hidden structure it wants to converge with. That’s exactly where classic ML starts to flinch, but people who think structurally — like you — start asking the right questions.

And totally feel you on EBMs. They’re a bit of a pain in terms of acceleration, but that expressive power is hard to beat. It’s clear you’re optimizing for meaning, not just speed — and that kind of design philosophy is rare.

If you're ever curious to experiment with frameworks that play at the boundary of emergence and symbolic reasoning, I’ve been building one behind the scenes. Still early days, but aligned with a lot of what you’re intuitively grasping here. Happy to share more in DMs if you’re ever in the mood to jam 🔁🧠

Either way, excited to see what you build. You’ve definitely got that upstream thinker energy.

[ruby dagger]

Hii Kaggle Community 👋🏻
Please checkout my latest datasets 👇🏻
https://www.kaggle.com/datasets/urvishahir/electric-vehicle-specifications-dataset-2025

[wicked berry]

What are the most 3 domains in Machine Learning that are high in demand in 2025 and more likely to stay high in demand in the future?

[agile cave]

Two different questions. The latter you can only figure out by actually coding and finding the Holes in the all the Hype of the latest thing. Grab a coffee.

know your algorithms

[lean compass]

Hi kaggle community, please anyone in the field of reinforcement learning, kindly let's chat. I seek collaboration on a project towards a hackathon.

[mystic sinew]

<@&1303433601177751593>

[uneven tulip]

guys phone no varifications problem😢

[little sage]

When training with an Nvidia GPU, add this to the top of your Jupyter notebooks so you know what you trained your model with.

I had to add this after these amazing GPU rental startups came along. So many options. I can dev with a 4090 then do a final training round with an H100, so this helps remember what I'm looking at when I read it later on.

# Check if torch can access the GPU
import torch
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
print(f"Using device: {device}")

# Run `nvidia-smi` to check GPU status
import subprocess
def check_nvidia_smi():
    try:
        result = subprocess.run(['nvidia-smi'], stdout=subprocess.PIPE, stderr=subprocess.PIPE, text=True)
        if result.returncode == 0:
            print("nvidia-smi output:")
            print(result.stdout)
        else:
            print("Error running nvidia-smi:")
            print(result.stderr)
    except FileNotFoundError:
        print("nvidia-smi command not found. Ensure NVIDIA drivers are installed.")
check_nvidia_smi()

P.S. My favorite VS Code Theme is Cyberpunk 2077, but the Genshin Impact theme is also really beautiful.

The doctors diagnosed me with schizophrenia, but I promise you he's real. He blew my brains out. Funniest guy I've ever met. Best dad jokes ever.

[slate mist]

Hello everyone, I'm currently working in a project ( Decoding SSVEP and Motor Imagery EEG signals for control applications)
Any recourses maybe useful to deep understand this topic would be helpful, 🙏🏻

[lethal otter]

<@&1303433601177751593> ya know theres bot scams in like every channel now right

[wide ridge]

Hello, I hope you're doing well.
If possible, I would like to know what a beginner in machine learning can start with as a practical project, and how ? .
Thank you in advance.😇

[little sage]

Bro, they hired me to build a machine-learning model to vet research papers for an immunology prize. Apparently the doctors who serve as judges are too expensive and so they want a model to do a sort of inital vetting process.

I got 64% accuracy so far and they're like "We want 90% accuracy". Bro... Lol. You want me to replace a doctor vetting these papers? Bro... You serious?

[little sage]

Bro! It's actually not that bad.

It's a regresion model. I embed the title, the entire abstract, and the first 50 sentences of the abstract using VoyageAi's embedding model. Then it's just a good-ol' neural network in torch.

I think with a bit more work I can get up to 70% accuracy.

[mint lark]

Hey i want to learn machine learning from where i can learn. Any suggestion

[wicked berry]

https://cdn.discordapp.com/attachments/1384218127809904712/1386253657460314133/image.png?ex=685908d3&is=6857b753&hm=933518c6468c23e9f19c2df6b95bda1376ba2d1ed3eb9ddfe8194becfc2bfe24&
https://cdn.discordapp.com/attachments/1384218127809904712/1386253675185307658/image.png?ex=685908d7&is=6857b757&hm=ab95600aca86a088a863e5a6605cf2869864ef17108dc5622d3ba258ba9d8ded&
https://cdn.discordapp.com/attachments/1384218127809904712/1386254951054839829/image.png?ex=68590a07&is=6857b887&hm=de4e771269f2abcd8c8871864c954e0172eabf0b866237868c70282ab7dc08e2&

[digital finch]

Anyone here interested in IOAI?

[nova merlin]

-# Only if u bag multiple golds n take part in IAIO too

[digital finch]

I will go to IOAI

idt I am eligible for IAIO

[finite pine]

I am trying to submit a notebook

But getting error

Always get an error of
install_requirements.sh:No such file or Directory

What to do I even added my PIP requirements in Dependices

[nova merlin]

Do cold mail the chairman once again the registration started for it too

Shame on u either if acm didn't register India or u didn't make it there in teams there

U already runner up for this team bro

[digital finch]

ACM India will start work on it shortly

they are too busy for IOAI

[digital finch]

There are two teams

[nova merlin]

IDC wanna see u in IAIO team with gold only

[digital finch]

I am not eligible for IAIO

[nova merlin]

Man everyone else this year performed so awfully

[digital finch]

who?

what?

[nova merlin]

For ioi n ioitc

Also expert

[nova merlin]

We 'd great expectations with the tc makers last year they will perform far better but every single guy fucked up except the real cp-ers

[orchid zodiac]

Been having some problems with the submission of the Prediction interval competition II: House price eventhough my submission is being generated at the end when i try to save my version it shows that there is no output can anyone help me out why is it like this

[cobalt shadow]

can someone help me make a submission for this competition? https://www.kaggle.com/c/store-sales-time-series-forecasting/data

i have already created a good model on the train dataset but i dont understand how i should use the test data to make a submission. please ping me if you are able to help me

[drowsy granite]

Hello kagglers, I have 3 notebooks reached vote but didn't take bronze medal, what should I do?

[barren gyro]

I know this is a big ask, but is there anyone that has experience in tableae that would help a me (beginner) work through some ideas. I have a small simple data set. I'd love to share my screen and ask a few questions if anyone is available. 👉 👈

[finite pine]

GPU is not Working, I tried but nothing T4 GPU show 0%

[tired wolf]

Hlo. I am new to data science and kaggle. Is there anyone new too so that we can learn together

[small cipher]

🤖 Just dropped: June edition of "Agents All You Need"
The AI agent landscape is moving at breakneck speed, and this month's newsletter covers the developments that actually matter for your business:
✨ What's inside: • Latest breakthrough in autonomous digital workers • Real case studies from companies deploying AI agents at scale • The tools and frameworks you need to know about • Practical implementation strategies (not just hype)
The race to build AI agents that can handle complex business tasks is heating up. While everyone's talking about it, few are sharing the actionable insights you need to get started.
This is reader-supported writing focused on cutting through the noise and delivering what you need to stay ahead.
Ready to dive deeper? https://bit.ly/3HXqNf9
What's your biggest challenge with AI agents right now? Drop it below – always curious to hear what's keeping you up at night.

[latent cobalt]

🔬 AI Hardware Research - Need Your Input!
Hey everyone! I'm doing research on AI/ML hardware accessibility in India and would love to get insights from this community.
Quick backstory: GPU prices in India are brutal, cloud costs add up fast, and there might be a gap for more affordable AI-focused solutions.
3-min survey about:

Your current setup and pain points
Budget constraints and preferences
What you'd want in an ideal AI GPU
Cloud vs local experiences

Drop your thoughts in the survey: https://forms.gle/UHGh1kzK1p9uiJSb9
I'll share the results here once I have enough responses. Your input could help identify real solutions for our community! 🙏
Feel free to share with other AI folks who might have insight

[little sage]

I started building ML models back in 2016. Almost 10 years ago. I spent most of my career not understanding anything. Until one day I understood.

Back then, people heard "machine-learning" and thought I was crazy. But I kept going. I was building machine-learning models on the streets of Chicago. Now, I am free. Don't have a boss over me. And when one tries to gain control of me, I spit at them.

Don't give up.

[nova merlin]

What do u do for living sir

[inner stag]

Hello Everyone,

My name is Ghulam Haider. I lead a dedicated team of developers and automation experts who provide end-to-end tech support for startups and growing businesses. We operate like a remote tech agency ready to step in whenever support is needed.

We offer:
-- Workflow automation (CRM, operations, email)
-- Machine learning models, data pipelines, and dashboards
-- Custom backend tools, APIs, and scripting
-- Reliable on-demand support for technical challenges

If you need assistance now or in the future, I would be glad to connect and explore how we can support your goals.

Here is my portfolio for reference:
LinkedIn: https://www.linkedin.com/in/ghulam-haidery/
GitHub: https://github.com/ghulamhaider65?tab=repositories
Kaggle: https://www.kaggle.com/ghulamhiader

Looking forward to hearing from you.

Best regards,
Ghulam Haider
Freelance Tech Consultant & Team Lead

[halcyon ether]

Hello my name is diaa,
I’m an aspiring data analyst with a solid foundation in financial analysis, banking, and office administration. I hold a Master’s in International Political Economy and bring extensive experience in financial operations, risk assessment, account management, and compliance. Known for being detail-oriented and analytical, I’m now building strong expertise in data interpretation, visualization, and tools such as Excel, SQL, and Tableau. I’m passionate about transforming data into actionable insights that support strategic decision-making, enhance performance, and drive measurable impact across organizations.

Eager to connect, work, collaborate on notebooks or projects and learn more.

Here is my links:
https://www.kaggle.com/diaaalmomani
https://www.linkedin.com/in/diaa-almomani-211318350/
https://github.com/diaamom

[agile ferry]

hii

[simple mulch]

Heyy everyone
how to actually start building, like I had audited the Machine learning specialization (Andrew NGs ) course on Coursera so they , don't provide assessments with audited courses
How can I implement it and have hands on learning
any project will require more prereuisite knowledge than i currently have
And if I use chat gpt or ai bots for projects then , how will I learn

i need help , so please if anyone can guide me

[glass arrow]

hi 🙂 are there models considered to be good for balanced_accuracy_score training? ty!

[minor dust]

So im training a yolo v8 small model using the visdrone dataset. I get good results but what happens is that sometimes it mistakes a vehicle for truck etc. I need it to track the objects as good as possible so I can get their trajectory data to train LSTM.
Dataset currently has 10 classes, what I wonder is if I can combine them together ? Would that cause any problems ? Like its going to call every type of vehicle it sees, just a vehicle ?

[fair zinc]

Bro learn how to use flask (it is a micro framework which will serve as a backend) and streamlit (easy ui for your ml model project)

[wicked berry]

Customer Churn Prediction (Telecom) by Aliii – Achieved 84.91% accuracy using LightGBM and SMOTE
link https://www.kaggle.com/dsv/12313316

[halcyon ether]

Hello

[tepid flower]

Salam @everyone,
I hope you all are doing well!

I'm a Computer Science graduate, I have around 2 years of practical experience in the ML field. I'm currently working within the Data field (handling multiple tasks related to data, under the title of ML Engineer).

I'm currently exploring Kaggle's competitions to sharpen my skills by working on real-world problems. However, I have multiple of questions since I've never participated in a Kaggle competition before.
Therefore, I'd be grateful if someone can help in this case.

Looking forward to connect, learn, and contribute with you all!

[void hound]

Hi, I'm Marco from Lima, Peru. I've been rushing the Deep Learning and Machine Learning specialization from coursera (taught by Andrew NGs). I felt these courses very strong in theory but a bit weak in practice. I'm not sure if with that experience I already able to start with the basic competitions of Kaggle. I still don't feel like I know well any machine/deep learning framework, the specializations used Tensor Flow some times. Any recommendation before starting with Kaggle?

[agile ferry]

Hii Can u share ur some project.

[worn barn]

Anyone familiar with semantic segmentation by SWIM ? https://github.com/SwinTransformer/Swin-Transformer-Semantic-Segmentation/tree/main

Are the masks of size (H/4,W/4) where H,W is the size of input image?

[scenic pond]

@everyone Does anyone know a better alternative to Ngrok

[slender sapphire]

yea, a real server

[wraith leaf]

Hi guys! Abhishek here, aspiring dev!

[signal crag]

may your models never overfit

cause mine certainly did

for the object detection task i was assigned for

[prime umbra]

hello team i am facaing issue where in my it shows 90K+ compitions in fact i joind none and also in my profile it shows 30K+ hosted compitions

[strong ether]

May our models be never over-fitting

[frank cloak]

hey guys had a little question. The projects that i make for kaggle competition solutions, is it possible that I download and save my notebooks (along with the outputs) and use them to showcase my skills on my own portfolio/resume? I'm asking cus i know the data used in the competitions is usually confidential. But is it okay if i can use the notebooks to showcase my work?

[short wedge]

This was a bug that popped up, it's fixed now.

[prime umbra]

Glad i got scared

[short wedge]

Sorry about that 😅

[final ledge]

I'm not sure if I correctly linked my account, this is a check if my message sent lol

[lyric spindle]

Hello everyone,
I'm new here. My name is Krishna and I'm a beginner and enthusiast eager to learn.
Actually I have a few beginner-level doubts related to AI and machine learning. Can I ask them here or is there a specific channel for that?

[sonic moss]

hey

i was completing courses in kaggle. In Intro to Programming i am facing a issue

even though i have completed the Function exercise it shows incomplete.

any idea or suggestion how i can improve that?

[prime umbra]

run full notebook

[edgy dagger]

you might have missed some code blocks

[warm tundra]

Hey everyone, I am completely NEW to data science and kaggle. I am learning my First Map, and I am very confused on how to do the exercise. Could anybody tell me what i hsould do on the first step? Im decent at coding i just dont understand if i should start a new notebook and do the exercise or do it on the designated page (the photo)

nvm i think i g ot it

[final flare]

Hi everyone! I took a bit of time and scraped all the COMPLETE derived data from the Mars Perseverance rover (solar day 1 - 1379) and organized it all into folders by data measurement type! I needed to parse all the data for a project im doing and couldnt find the complete derived data set so I'm hoping the upload will be helpful for others!
https://www.kaggle.com/datasets/lolajackson/mars-2020-perseverance-meda-rover-data-derived/data

[astral ledge]

Hi Guys. I am a aspiring dev looking for entry lev roles. I guess most of the poeple here are pretty good experienced. If any one has any leads i really really !! appricate it

[prime umbra]

The red message in your notebook is not an error, it's just a warning:

It still works for now, but will be removed in a future version of GeoPandas.

You should not worry immediately, unless you're updating GeoPandas to version 1.0+ later.

[rugged wedge]

🚨 Collab Call: Gemma 3n + Education Hackers Wanted!

Hey folks — I’m Apreddy, solo builder of Lab2Life — a global hands-on science learning platform that just shook up Bolt Hackathon 2025. Now, I'm extending it to Gemma 3n (Kaggle Hackathon) to bring offline, privacy-first, multimodal science learning to underserved kids around the world.

Looking for 1 solid collaborator who’s got:
✅ Experience running Gemma 3n locally (preferably with Kaggle or on-device setup)
✅ Knowledge of multimodal AI (images, voice, text fusion)
✅ Comfort with Edge deployment, privacy-first AI, or education tech
✅ Bonus if you know lightweight backend/devops for packaging

💼 What You’ll Do:
Help build and demo a voice+vision AI Lab Assistant powered by Gemma 3n
Assist with integration (voice prompts, object recognition from webcam, etc.)

🎯 Goal: Build something powerful. Ethical. Fundable.
DM me or drop your GitHub/Kaggle/Portfolio if you're game.
Let’s bring science to every child — even offline.

[lyric spindle]

Hello, I have a doubt and would really appreciate some guidance.
I’m currently doing a Generative AI course where I’m learning about Machine Learning, Deep Learning, and different models. However, I’m a bit confused about what to pursue next, whether I should go for Data Science, Prompt Engineering, or something else, since my friends have different suggestions.
Any advice would be helpful.

[torn ledge]

Hi everyone my ai conversions has some error showing in https://huggingface.co/spaces/r3gm/Aesthetic_RVC_Inference_HF

Can anyone help me

Idk why I’m getting this error

[prime umbra]

first find one thing that you like

Do one thing at a time

and deciding to pursue Data Science, Prompt Engineering,or something else totally depends on your intrest

[wicked berry]

prompt engineering isnt real lmao

[ember helm]

hi folks im currently looking for a mentor for a university ML project would anyone guide me in this?

[green condor]

hi guys, i'm also looking for a mentor to guide through my university project .
I'm very passionate and worked on couple AI projects and did some courses on udemy before but now I need guidance to advance .

[prime umbra]

You can find guided projects in cognitive classs .ai website

[prime umbra]

.

[green condor]

thank you for answering !!
I specifically want to create a text-to-SQL model .
is there a course close to that ?

[ember helm]

like he said we need to find mentors that are currently working in the industry wait let me share

[prime umbra]

well you can find list in this link
https://cognitiveclass.ai/courses?type[]=guided_project

in your case this might not be helpful

[ember helm]

yeah, he was stubborn on this, he was like YOU HAVE TO FIND ONE YOURSELF

[lyric spindle]

Alright, thanks for the advice

[sonic moss]

yea got it resolved

[sonic moss]

that was the issue, ty

[visual prawn]

where to sign up 5-Day Gen AI Intensive Course with Google

[nova phoenix]

hi

[prime umbra]

https://www.kaggle.com/learn-guide/5-day-genai

[prime umbra]

lol

[ember helm]

So I'm basically cooked

[prime umbra]

pretty much

yes

[silk merlin]

hi

i just switched from codeforces to jaggle

[pulsar ferry]

hi joined to get some help in ml models

[green condor]

Me too .
I transitioned yesterday

[silk merlin]

yeh

it's time to realize cp is kinda useless in my university path

[agile ferry]

Damn!!

[halcyon wave]

hello guys, currently pursuing engineering from banasthali viidyapith and want to participate in titanic- ML from disaster hacakthon on kaggle if anyone interested come and join with me

[autumn stratus]

Hello, i want to learn more about ai/ml from kaggle

[cloud hedge]

Hii! I'm a Data Scientist passionate about transforming complex data into strategic insights that drive business decisions. I combine solid statistical foundations with modern ML tools to solve problems.