// Featured

Agent Harnesses Compared: Loops, Memory and Tool Calls

Oct 20, 2026 · 8 min read · 👁 – · ● AI Engineering

A practical breakdown of how popular agent frameworks handle looping, memory and tool calling, with real tradeoffs.

// Latest

We tested open-weight models across coding, reasoning and tool-use tasks. Results, cost and what the numbers mean.

Oct 14, 2026 · 6 min read · 👁 –

A curated, opinionated list of MCP servers that improve agent capabilities, security and developer experience.

Oct 8, 2026 · 5 min read · 👁 –

How prompt caching works, when it saves tokens and money, and how to use it in your agent workflows.

Oct 1, 2026 · 4 min read · 👁 –

Reliable tool calling patterns that reduce hallucinations and improve agent reliability.

Sep 24, 2026 · 7 min read · 👁 –

Designing memory for agents that run for hours or days: persistence, summarization and eviction.

Sep 15, 2026 · 6 min read · 👁 –