agno

After reading hundreds of papers on agentic memory and trying out every possible tool, I came to the simple conclusion that maybe we're looking at memory wrong. Memory is just... learning. Learning about the user, the task at hand, learning insights and patterns, learning from decisions - good and bad, the feedback received. Learning from every interaction. Everything else is integration (how the agent uses these learnings) and curation (decay, pruning, deduplication). So I built Learning Machines: A system that helps agents continuously learn from every interaction. I started working on it dec 31, and got a basic working version yesterday. Here's the PR for those interested: [learning-machine-v0](https://github.com/agno-agi/agno/pull/5897) This post digs into the technical details.

Hi HN, Excited to share Agno, a framework and runtime for multi-agent systems. Think of it as FastAPI for AI Agents. At its core is the AgentOS, a high-performance server/runtime that helps you run and manage AI agents, multi-agent teams, and step-based agentic workflows — all inside your own cloud, with full privacy and no external data sharing. What makes it different • Fast & lightweight — Agents instantiate in ~3μs and use ~6.6 KiB of memory on average (tested on M4 MacBook Pro). • Runtime architecture — Async, stateless, horizontally scalable runtime built on FastAPI. • Integrated UI — Test, monitor, and manage your agents and teams in real time. • Private by design — Runs entirely in your environment. No vendor lock-in, no telemetry, no external tracing. A key distinction: the control plane connects directly to your AgentOS runtime

Thank you for the feedback and the kind words. Agree that the cookbooks have gotten messy. Not an excuse but sharing the root case behind it: we're building very, very fast and putting examples out for users quickly. We maintain backwards compatibility so sometimes you see 2 examples doing the same thing. I'll make it a point to clean up the cookbooks and share more examples under this comment. Here are 2 to get started: - Content creator team: https://github.com/agno-agi/agno/blob/main/cookbook/examples... - Blog post generator workflow: https://github.com/agno-agi/agno/blob/main/cookbook/workflow... Both are easily extensible. Always available for feedback at ashpreet[at]agno[dot]com

Good point. The cookbook can be hard to navigate right now, but that's mostly because the team is putting out a tremendous amount of work and updating things constantly, which is a good problem to have. This example might be close to what you're describing: https://github.com/agno-agi/agno/blob/main/cookbook/workflow... It chains agents for web research, content extraction, and writing with citations. I used it as a starting point for a couple projects that are now in production. It helped clarify how to structure workflows.

It's why these no-code/vibe-code solutions like bolt, lovable, and replit are great at hackathons, demos, or basic front-ends but there's a giant cliff past there. Scroll through things like https://www.yourware.so/ which is a no-code gallery of apps. There's this utility threshold due to a 1967 observation by Melvin Conway: > [O]rganizations which design systems (in the broad sense used here) are constrained to produce designs which are copies of the communication structures of these organizations. https://en.wikipedia.org/wiki/Conway%27s_law The next step only comes from the next structure. Lovable's multiplayer mode (https://lovable.dev/blog/lovable-2-0) combined with Agno teams (https://github.com/agno-agi/agno) might be a suitable solution if you can define the roles right. Some can be non or "semi"-human (if you can get the dynamic