low pewter
Howdy OpenAI Devs,
I'm building new data storage and pipeline services for the budding AI ecosystem.
One of the challenges I faced when creating a fine tuned ChatGPT model was to automatically 'save my work' outside the OpenAI cloud platform, in addition to being able to easily generate new JSONL training files that I could then upload to further train my models.
That's the problem I aim to solve with my new project, HighContext: https://www.highcontext.ai/integrations
With just a few clicks, you get hourly backups, automatic parsing of your JSONL training files into individual JSON blobs, re-packaging sets of JSON blobs into JSONL training files, redundancy in the event of OpenAI outages, plus more. Please give it a try and let me know what you think!
P.S. What additional data challenges are you facing while building on OpenAI's base models? Let me know, I'd be curious to assist however I can.
jagged mountain