#ai-agents

2 posts • 1 shared link

Articles & Writings

File-Based Memory Is a Terrible Idea That Somehow Works

A production diary of tiered episodic memory in AI agents. Three markdown files, an SQLite database, and a lobster that somehow remembers what you said three days ago.

February 28, 2026 • 15 min read

Read article

ai-agentsopenclawmemoryproduction

Stop Renting Your Workflow: Building a Custom AI Coding Agent with Pi

Four tools, a settings file, and full control - how I built my own AI coding agent with Pi instead of paying $200/month for a CLI that keeps changing.

February 11, 2026 • 11 min read

Read article

ai-agentscoding-toolsllmdeveloper-toolspi-coding-agent

Shared Links & Reads

April 15, 2026 example.com

What we learned from a year of building LLM agents

llmai-agents

A practical review of building agent workflows in production:

Multi-agent orchestrations are rarely needed; start with simple linear chains.
LLM evaluation pipelines are the single most important factor for success.
Focus on latency and caching early on.