lowkey tired of being a magnet for karma farmers with hot takes. Nuance should be celebrated. GitHub is working hard on dealing with the sheer volume of code that agentic software engineering produces. And at the same time, it is incredibly frustrating to all our users and customers that we aren't there yet. And at the same time, we would be derelict in our duty to developers if we weren't sweating to understand the future shape of things. All of these can be true at once

GOOD NEWS: The 3 MB free and 10 MB paid compressed Worker size caps are gone! Cloudflare now checks only uncompressed bundle size, up to 64 MiB on all plans. Check yours with wrangler deploy --dry-run: Total Upload is what counts. 🔥

Have been following @DhravyaShah's journey with @supermemory for the last 2 years or so, and it's been nothing short of remarkable: > solo founder, lean team > built the best-in-class memory infra for agents > ~6 fig monthly revenue, growing at breakneck pace Honored to be powering their affiliate program with @dubdotco →

+1 for "context engineering" over "prompt engineering". People associate prompts with short task descriptions you'd give an LLM in your day-to-day use. When in every industrial-strength LLM app, context engineering is the delicate art and science of filling the context window with just the right information for the next step. Science because doing this right involves task descriptions and explanations, few shot examples, RAG, related (possibly multimodal) data, tools, state and history, compacting... Too little or of the wrong form and the LLM doesn't have the right context for optimal performance. Too much or too irrelevant and the LLM costs might go up and performance might come down. Doing this well is highly non-trivial. And art because of the guiding intuition around LLM psychology of people spirits. On top of context engineering itself, an LLM app has to: - break up problems just right into control flows - pack the context windows just right - dispatch calls to LLMs of the right kind and capability - handle generation-verification UIUX flows - a lot more - guardrails, security, evals, parallelism, prefetching, ... So context engineering is just one small piece of an emerging thick layer of non-trivial software that coordinates individual LLM calls (and a lot more) into full LLM apps. The term "ChatGPT wrapper" is tired and really, really wrong.

im looking for a co-founder: >building together is more fun >im in SF working from @fdotinc (they invested) we'll get along well if: >you're an overachiever >but chill >love simulations we're gonna build a multi-agent, fully emergent life sim game. dms open. (you can actually see @fdotinc in this GIF)

I'm going against Grok Bot excited to announce it gives you AI teammates that connect to all your tools, understand your context, and take action to get your work done on schedule