Author
Name
Daniel Botha
Image by
Annie Ruygt
Building Agents that Don't Break Themselves
Where your agent runs and where it runs code are two different questions. Answer them separately.
1,363 words~6 min read
Related reading
Break Your Agent on Purpose: A Failure-Injection Sandbox for Tool Boundaries
Last week a post on DEV asked a question I can't stop thinking about: we keep handing AI agents more...
Agent Memory, Part 2: What I Built Broke
In part 1 I wrote about deciding to dig agent memory to the bottom. This part is about building. It...
What you actually need to ship an AI agent
Everyone's building agents. Half of them are running. The other half have "active plans." I've been...
Agents Should Be Durable, Not Long-Lived
Separate durable agent state from short-lived compute workers to make long-running AI systems cheaper, safer, and easier to scale.
Why agent projects stall after the demo👨🔧
Build agents that survive production
You Don't Need a Better Agent. You Need a Better Debug Loop.
You don't need a better agent. You need a better debug loop. I've been building agentic systems long...
