We use Obsidian across our team, have one primary vault that we use for all public docs and some team mates have a private vault alongside it. Works decently well including bases for databases.
I dislike writing docs in Obsidian so I'm constantly on the edge of changing... but to what???
@hwchase17@cognition@FactoryAI@karpathy I still wonder how we weight the content in a way that is useful for the LLM. In my explorations it seems to find it difficult to know what’s important
Everyone is writing about agent loops right now. Including us at Cursor, because they're so powerful. But here's a prediction: a year from now, nobody will be talking about them.
Not because they weren't useful. Because they'll work right out of the box. Batteries included. No instructions necessary.
Feels a lot like prompt engineering two years ago. It was incredibly important. People wrote courses on it. Now you just talk to your agent like a normal person.
That's the strange thing about AI right now. You learn something critical, get huge gains, and before long it's the new normal and something else is the bottleneck. So the alpha isn't what you know. It's how fast you learn it, and how easily you can let it go.
💯
It feels more clear than ever though that founders (like me) need to keep building...
There's more possibility to be harnessed by technology, opportunity to make impact and if we sit on the sidelines while things settle we'll be irrelevant in 2-3 years.
Wild times but the endless optimists need to lean in!
@Alfred_Lin@paulg how much is direction emergent? and the key is ability to ship quickly and also incorporate learnings?
winning = ship lots in right direction = ship lots + listen to customers
AI SDK 7 is now available.
Introducing: reasoning control, agent-level tool approval, tool and runtime context, file and skill uploads, MCP Apps, durable workflows, terminal UI, sandbox support, harness integrations, telemetry, lifecycle events, and more.
Most agentic software dies in the gap between what you hoped it would be and what your users actually value.
@run_otis lives in that gap: it learns what people really want, and turns it into the product.
Turning ideas into what others actually value. That's the whole thing.
More soon.