Skip to content
Initialising environment000

Blog · Coming soon

Notes from the lab.

Write-ups on design, reinforcement learning, AI agents and the engineering around them: what we try, what works, and what we'd do differently. The first posts are being written.

Talk to us meanwhile
Scroll to explore

What we'll write about

Four threads, first posts in progress.

  1. Thread 01 / 04

    Design and frontend

    Interfaces, design systems, motion and performance: building sites and products people enjoy using.

  2. Thread 02 / 04

    Reinforcement learning in practice

    Environments, rewards and evaluation: the parts that decide whether a policy survives contact with the real world.

  3. Thread 03 / 04

    Agents that can be trusted

    Planning, tool use, guardrails and the evaluations that tell you an agent is ready.

  4. Thread 04 / 04

    Engineering notes

    The software around the models: data pipelines, infrastructure, apps and the handover.

Want the long version now? Ask us.

Start a project
Next · Get in touchContact