Ask anything about AI agents.
Grounded in 51 field notes, research, and every episode
Concise, sourced patterns drawn from research and production practice.
Big agent-engineering ideas unpacked without the hype or hand-waving.
Query the whole knowledge base and get a direct, grounded answer.
Read → act → observe → correct. A durable mental model for every agent.
From the knowledge base
High-signal ideas for building agents that are capable, legible, and reliable.
An agent loop repeatedly reads the current state, chooses an action, uses a tool, observes the result, and corrects its plan. The model is only one component: reliable agents also need…
UK AI Security Institute evaluations found that increasing an agent's token budget from 1 million to 10 million tokens improved success on software-engineering tasks by roughly 25 percent.…
Tool-using agents work best with narrow tools, explicit schemas, actionable error messages, and observable results. Validate arguments before execution, use idempotency keys for side…
Short-term memory is the working context supplied on each model call: recent messages, the current plan, tool results, and task-local state; it is bounded by the context window and…
Agent evaluations should separate model quality from scaffold, tool, and environment failures. Single-run pass rates hide nondeterminism, while overly short time or token limits can…
Use multiple agents when subtasks are genuinely independent, require distinct tools or context, or benefit from parallel search or review; a single agent is usually cheaper, faster, and…
Stay in the loop
Explore the notes, follow the source material, or ask the knowledge base about the problem on your desk.