You’re too kind. 🙏 I don’t know how much value there is in ideas without execution but I’m happy to give it a go! Here’s one — and it’s quite meta, because it’s about what breaks when AI slop makes certain things too cheap. If anyone would like to build on this, I’d be happy to (1) be an adviser (2) take it as a signal to clean up and publish more ideas. https://t.co/ZUxswR3emX
Announcing OpenWorker! An open-source agent that doesn't just chat with you, but delivers finished work -- like hand you a polished document, send a slack message, or update a calendar entry.
Ask it to prepare a customer brief, untangle your calendar, draft a report, or triage a Slack alert. It works across your files and everyday tools, produces the deliverable, and checks in before doing anything consequential.
OpenWorker runs on your Mac, with Windows support coming soon. It does not lock you into any one model. Bring your own API key and run it with GPT 5.6 Sol, Claude Fable, Gemini 3.6, an open weight model (like Kimi, GLM, DeepSeek, Inkling), or Ollama to keep your data local. Your data does not leave your machine except through an LLM provider and integrations that you choose.
@rohitcprasad and I are building OpenWorker because AI coworkers are an important way to get work done, and we want there to be an open, privacy-preserving, model-independent option. Check it out and let us know what you think!
Try it out: https://t.co/P0mGnI1o31 (requires your own API key)
Source code: https://t.co/NYCiTD6hSq
Underrated - “The best researchers and engineers develop an intuition for where the signal is likely to be found long before they possess complete information.”
@sama after a long coding conversation with Canvas, the browser becomes unresponsive. And forking conversation also doesn’t solve the issue. Feature request : reference another conversation to continue. Thanks 🙏
@jeremyphoward@Yuchenj_UW I believe ULMFiT doesn’t get enough credit for being the first to achieve transfer learning for NLP.
ULMFiT is where I started my DeepLearning journey, my first ever model to get to production 🙂
Thank you.
@sama chatgpt is not able to fork a conversation. And still thinks it can, so it’s replying with empty links. This would be a great feature addition, for things like a project to study a subject in depth.
Thanks!
@karpathy Nice!
Curious, how much of your software engineering experience you had to use to achieve this? (For writing the prompts, building the app bundle, testing)
Not criticizing, just curious.
@paulg @HistoryEJ A lot of folks who grew up in countries where english was not spoken actually use a lot of "written-first" words in their writing because that is how they learned the language.
@whovivekshukla@gaganbiyani Two years back I would have said join a FAANG in India (that’s what I did, and now I’m in Bay Area).
Things have changed since the layoffs, but things should go back to what it was in a year or two.
@PranotiNagarkar@rishikum@rotimatic Device replacement is taking forever, every few days customer support has been promising to ship the device in next 48 hours. It’s already been 26 days since the old device reached your facility. I still haven’t received order number.
I feel a bit out of touch with current fashions, because a couple days ago an Indian founder in the current batch asked if he should base the company in SF or India, and I said the Indian startup community seemed to be thriving and it would be fine if he wanted to go back.