LLMs and Agents

Posts

ML Environment Engineering: Building Machines That Build Machines

Instead of building models, build the environment in which models get built. How automated ML agents can run experiment loops against strong evaluation harnesses — and where humans still need to step in.

Read post →

Agents: From Inference-Time Scaffolding to Inference-Time Compute

How reasoning and autonomy moved from external scaffolding, to imitation learning, to self-improvement — arriving at models that perform these behaviors natively within their own reasoning trace.

Read post →