/ About this event
Building Self-Improving AI Agents: From Loop to Deployment
AI is moving past simple chatbots toward autonomous agents capable of complex reasoning. The next step is agents that learn from feedback and improve their own performance over time.
Hosted by GRAIL, this session explores recursive self-improving (RSI) agents. This two-part event features a panel of academic and industry leaders followed by a hands-on technical workshop.
During the workshop, participants will build on GRAIL's OpenClaw-based infrastructure called ApexClaw, so you don't need your own agent infrastructure to take part.
What You'll Learn #The Future of Agents: A panel discussion with academic and industry leaders on where self-improving systems are heading. #The Agentic Loop: How agents handle tool use, memory, evals, and automated improvement. #Introduction to ApexClaw: How to go beyond general-purpose LLMs with a framework built for agents. #Live Implementation: A guided session on prototyping your first self-improving agent, whether that is an AI employee for your startup, a research agent for your lab, a coding copilot that rewrites its own workflows, or an operations agent that gets sharper every time it runs.
Who Should Attend #Founders looking to build AI employees or operations agents for their startup. #Researchers and PhD Students interested in research agents for their lab. #Engineers and Developers exploring agent loops, evals, and OpenClaw. #Investors interested in the agentic AI space.
What Participants Receive All attendees will receive exclusive free credits for ApexClaw, allowing you to implement your own agent during the workshop and continue building your own digital researcher after the event concludes.
The location will be provided to registered attendees.
Duration 2-hour
This event is a part of #SFTechWeek—a week of events hosted by VCs and startups to bring together the tech ecosystem. Learn more at www.tech-week.com.
