Local Models & Self-Evolving Harnesses

Presentation Link#

The seventy-first MeetUp of the Machine Learning Singapore Group, was titled : “MLSG - The Agent Harness Revolution”, during which we also welcomed back Sam Witteveen from a two-month vacation in San Francisco.

My Presentation#

My talk was titled “Local Models & Self-Evolving Harnesses”. On top of my typical research oriented takes, I took the opportunity to reflect on why it is essential for individuals and companies to “Own Your AI”. Overall, I used the following outline :

• Basic Harnesses

• Self-Modifying Harnesses

• Token Sourcing Risks

  • … and mitigation

• Wrap-up & QR-code

Slides#

The slides for my talk, which contain links to all of the reference materials and sources, are here :

If there are any questions about the presentation please ask below, or contact me using the details given on the slides themselves.

Other Presentations#

The MeetUp also featured:

Sam Witteveen giving his talk (previously delivered at Google Cloud Next in Las Vegas) “Patterns of Successful Agent Harnesses”. In the talk Sam focused on the core patterns and components of successful agents and harnesses, covering patterns from coding agents like Claude Code & Codex through personal agents like OpenClaw & Hermes. These patterns clearly show what is working in building modern agents.

Raghav Bhardwaj gave a lightning talk “Trying to download more RAM for my life”, about his experiences hacking OpenClaw to combine 100s of projects involving chats, files, tasks, tools, and memories into one working loop.

Acknowledgements#

Many thanks to the Google team, who allowed us to use Google’s Developer Space, and generously provided food for our audience.