Follow AI Builders — Not Influencers

Today @AAAzzam will walk through how to setup agent sandboxes in @modal , my favorite compute platform Modal has wicked fast cold start times. The devex makes remote compute feel local Link: https://maven.com/p/0684ab/don-t-build-agents-build-environments-instead notes/recording sent to people who sign up

There’s a really good book detailing all of this kind of hedge fund trading behavior that I read. I highly recommend it because it chronicles all the various beliefs and results around those beliefs. The moral of the story is: having these hardened beliefs / thesis (algo, gut-based, etc) ultimately leads to ruin and the world is very complicated!

RT LlamaIndex 🦙 Not every page in your PDF needs the same treatment. A scanned cover, a dense table, a clean text page, a figure-heavy diagram.... most parsing pipelines throw all of them at the same parser, forcing a tradeoff between cost, speed, and accuracy across the whole document. We built 𝗣𝗮𝗿𝘀𝗲 𝗚𝗮𝘁𝗲𝘄𝗮𝘆 to fix that. It uses LiteParse's 𝘪𝘴_𝘤𝘰𝘮𝘱𝘭𝘦𝘹 functionality to estimate complexity page-by-page and routes each page to the right tier based on why it's complex (scanned, sparse text, garbled encoding, vector text, embedded images) and how severely. Simple pages get parsed for free, in-process. Genuinely hard pages get routed to more capable LlamaParse tiers. No manual heuristics, no flat one-size-fits-all pipeline. And it's also available as an MCP server, so your agents can estimate complexity and choose a parsing tier themselves! The result: you stop paying premium prices for pages that never needed it, without sacrificing accuracy on the ones that do. 📚️ Full writeup: https://www.llamaindex.ai/blog/parse-gateway-smart-page-level-document-parser-routing 👩‍💻 Try it: https://github.com/run-llama/parse-gateway

Ok so I’m starting to wrap my head around AI agents for knowledge work teams. I’m realizing everyone will want an app like Codex/GPT work for individual use, as well as an agent they can use collaboratively. What the 2nd one looks like is unclear. Every company is different in size, the way they communicate, and what they do. It makes it hard to create a one size fits all “agent solution”… Setting up slack agents is hard but it’s not the hardest part. It’s the maintenance, permissions and the onboarding of new team members so they understand how to use the company agent. Also evals, models, skills, connections, channels, internal docs, versions, bugs and like 50 other things. If you’re a company that has a great setup for agents that your whole team uses, I would love to hear about it.

TIL. how the heck did linkedin beat twitter to this feature??

RT Adam Holter Jason is great! one of my favorite OpenAI employees. Tibo fuels my addiction, so I'm more dependent on him, but Jason is definitely the most fun. Y'all should go watch his AI Engineer World's Fair Summit talk. It was really good: https://www.youtube.com/watch?v=il1c1a2FufU

RT Hamza Jason is the most creative person OpenAI might have (or at least we can interact with) This video is really beautiful and nice, although of course it is a marketing video and there is no denying about it but this is really well done and shows how ChatGPT/Codex can actually benefit humans