AI Agents
Building useful AI agents is mostly systems work: choosing the right tools, supplying durable context, controlling the execution loop, and verifying outcomes. These guides focus on the engineering patterns that turn a model into a dependable software system.
11 guidesLatest 31 Oct 2026
Start here
- 1Structuring tools for AI AgentsStart with the interface: shape tools around the questions an agent needs answered.AI Agents, Structured Generation
- 2How to Build an MCP ServerPut a small set of tools behind a real server and test it from an MCP client.AI Agents, LLM Infrastructure
- 3Plain-Text Context Files for LLMsGive the agent instructions it can load at the start of every session.AI Agents
- 4Loop / Harness EngineeringSee how tools, context and verification fit together in a production loop.AI Agents, LLM Infrastructure
- 5The Tool List Is Not an Authorization BoundaryDecide what the loop may do, and enforce it where each tool executes.AI Agents, LLM Infrastructure
More guides
- The Temptation to Build EverythingBuilding a tool is now quicker and more fun than adopting one, which is fine until other people depend on it.AI Agents
- Your Agent Memory Needs Two ClocksA single timestamp cannot distinguish a late correction from a real-time state change.AI Agents, LLM Infrastructure
- Delete System-Prompt Rules One Regression at a TimeTie every prompt workaround to a failure case, then test its removal when the model changes.AI Agents, LLM Infrastructure
- Treat Reasoning Controls Like API MigrationsShip the model, reasoning mode, and history adapter as one versioned protocol.AI Agents, LLM Infrastructure
- What Pass Rates Hide About Long-Horizon AgentsReport the share of runs that reach each milestone, so you can see the step where an agent starts failing.AI Agents
- Speculative Agents Need a Commit BoundarySpeculative execution cuts agent latency only when prediction and commitment are separate operations.AI Agents, LLM Infrastructure