SPEAKER
Lamis Mukta
Member of Technical Staff · Anthropic
I'm a Member of Technical Staff at Anthropic, working on our Applied AI team with founders and start-ups who are building at the frontier and pushing the boundary of what's possible with our models. I've previously held engineering, product, and research roles at early stage start-ups, been a founder myself, and worked as a maths researcher at Oxford prior to that.
Learning while you sleep: Beyond memory to dreamingOn demand
1 June, 4:55 PM · 30 min Conference Session
Past sessions
See complete program1 June
4:55 PM · 30 min
Learning while you sleep: Beyond memory to dreaming
Early implementations of memory were key to managing context and paving the way toward continual learning. We'll walk through the bottlenecks we're seeing with this approach today, and introduce the concept of dreaming: a second derivative process which periodically prunes and curates memory and leads to measurably better agent performance over time. This will include an architectural overview of designing a dreaming service in practice.
2 June
2:30 PM · 100 min
Ship your first managed agent
Introducing Claude Managed Agents: a pre-built, configurable agent harness that runs in managed infrastructure. In this session, you'll build and ship a working incident-investigator agent on our Managed Agents platform: define an Agent, Environment, and Session, stream events, and wire up custom tools, all in six functions. You'll leave with a running agent, the mental model for the server-side loop, and a roadmap to production features like subagents, vaults, and memory.
