NG 2026 can't get a job, am I cooked

[hexed hinge]

Tyty

I never know what like fits in a skills section

Haven't kept count really, 5-10 a day

[astral hollow]

Great resume

[rare vine]

holy havent seen such a non ML and organic resume in a while

[tardy raven]

What type of jobs are you looking for? Embedded or general sde?

[hexed hinge]

Moreso embedded/robotics, but open to regular swe

[hexed hinge]

Thank you!

[white canopy]

try to look for a combination of both

[warped beacon]

lowk put role in italics and like princeton in bold. its a big name make it stand out

[restive moat]

^ and swap title and position

[hexed hinge]

Can do

Also i need advice. The NBA project is actually bigger than i have on my resume, i'm continuing my research w a prof here in hopes of pub

Ive made more bullet points for it but like

Idk if to put it

Its part clustering & archetype analysis and part performance analysis

[tardy raven]

I personally think you should put more bullet points in work experiences and reduce the project experiences you have. Once you have industry experiences recruiters rarely look at projects… unless it’s something really impressive

[hexed hinge]

I see i see, i wanted to portray more diversity in work but maybe i should just expand my internships

[broken jolt]

yeah no, experience is king

I am only ever asked about my top 2 experiences, once I was asked about my 3rd, but never ever have i been asked about a project

Also, u still have a semester left maybe try to do some robotics research on the side for extra relevant xp

[hexed hinge]

I should

[hexed hinge]

Hmm ive been asked abt some robotics projects before but yea def more internship

Should i add my research as a position?

Bullet points would be something like:

• Built a Python data pipeline (Selenium, BeautifulSoup, SQLite) that scraped 1,247 NBA games and processed 55,382 L2M events into a 4-table relational database with fully automated ingestion.
• Implemented ETL + feature engineering in Python/pandas to normalize 14 call-type categories and compute accuracy metrics for 218 referees.
• Developed an ML workflow (scikit-learn) using 9 engineered features and 55k labeled samples, producing 30 analytics outputs via custom matplotlib tools.
• Scraped and parsed ~430 Basketball-Reference referee pages using Python/BeautifulSoup.
• Computed 15+ referee metrics and ran LOF, K-Means (k=4), PCA, t-SNE across all referees, generating automated outlier detection and cluster visualizations with scikit-learn and matplotlib.
• Integrated NBA API play-by-play feeds for 2,000+ games, cleaning ~15–20k events/game and calculating in-season foul statistics (techs, clutch fouls, home/away splits).

Obviously need to be shortened, but essentially the project is two parts, analysis of referee performance and clustering of referee behaviors