Post

Going to attempt to do more chill coverage right now of the AGNTCon + MCPCon morning keynotes because I’m sleepy, but because it’s dark in here these threads are also my notes for myself (I’m back to doing hand-written notes because LLMs can read my handwriting better than I can)

Lots of awards being given out right now for contributing to the wider community. Congrats to the winners, I couldn’t keep up with them all. A small sampling: Agentic Tokyo, solo.io, arcade.dev, Pydantic (individuals from the latter two for their contributions to OSS here)

And the keystone award winner is…🥁 Den Delimarsky of @anthropic.com and lead maintainer of MCP

Next up is Jonathan Bryce, on moving to better systems for AI with a focus on PyTorch. “Loops and feedback need to move all the way back down the stack” We need to move beyond “brute-force” - we still able to wring more capability out of smaller models!

Jonathan shares some history of model building focus, starting with pretraining, then post-training, and now “on the job” training - retraining models regularly with feedback from agents. This isn’t just about LLMs, but this shows the importance of continual learning. Right now the loops aren’t

truly continual (mostly), but if they were? Workflows would be vastly simplified. Shopify: AI judges on specific tasks that their models do, if they don’t outperform humans or frontier models, retrain on new data.

The PyTorch Foundation’s goal is to build models on any chip, any cloud, with any agent. Now Chi Wang formerly of Google DeepMind and now at AG2 to speak about how we can get personal agents to work together.

The “new network” consists of members (people and agents) sharing rooms with their own contexts, with agents able to also work directly together. But agents need accountability and rules to safeguard data.

A theme I’ve seen in a few talks about agent-agent communication is that agents can carry requests from its owners but never their authority. Wang proposes a broker in between agents to do this.

This shows an example workflow for task-focused agents coordinating to report a bug, implement a fix, and approve the changes.

“Coordination could change the scale at which people organize work”

Now we are hearing from Paul Conyngham, the guy who worked with multiple scientists to develop personalized mRNA cancer vaccines for his dog, and is scaling that to develop computationally designed mRNA cancer vaccines.

Already, he is speaking of building ontologies and doing things we’ve done in healthtech and biotech for doing this work - a good thing not covered in any of the popular coverage of what he and his team did. An early idea was “finding the diff” between baseline DNA and cancer cells. Not a new idea,

But he outlines the technologies that converged to allow this to happen: individual sequencing, computational genomics and AI, and mRNA vaccines. He explains that he used AI as a learning partner to learn how these technologies worked and the science behind them.

Hard to argue with results. His new venture Gamgee, wants to scale this up eventually to humans.

The real thrust of this talk is how to work effectively across disciplines. Emphasizes that the human is the one that has to “be the agent” to make it happen.

Now @ashtom.bsky.social is up to talk about the future of version control for agents, something I know a lot of people here are interested in and working on. He opens to show us how number of PRs is skyrocketing over time and the size of PRs is 3x what it was 18 months ago, causing more changes