Guides

Implementation patterns, technical comparisons and decision rules for engineers building with language models and agents.

15 guidesLatest 31 Oct 2026

2026

31 OctThe 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
26 JulYour Agent Memory Needs Two ClocksA single timestamp cannot distinguish a late correction from a real-time state change.AI Agents, LLM Infrastructure
26 JulDelete 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
25 JulTreat Reasoning Controls Like API MigrationsShip the model, reasoning mode, and history adapter as one versioned protocol.AI Agents, LLM Infrastructure
25 JulThe Tool List Is Not an Authorization BoundaryRemoving a tool from the model's list doesn't revoke the credentials behind it, so check permissions when the call executes.AI Agents, LLM Infrastructure
25 JulWhat 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
11 JulSliding-Window Prefix Caches Need Physical ProofIn hybrid sliding-window models, the prefix tree can remember tokens whose local KV slots are already gone.LLM Infrastructure
11 JulSpeculative Agents Need a Commit BoundarySpeculative execution cuts agent latency only when prediction and commitment are separate operations.AI Agents, LLM Infrastructure
28 JunLoop / Harness EngineeringAgent loops turn prompts into one part of a production harness: tools, state, checks, approvals, and exit conditions.AI Agents, LLM Infrastructure

2025

2024