AI agent learning beats demo flashiness
Abundance mindset with AI SRE competitors
The coding interview needs to die
Two years in America
Hallucination rate is the wrong question
Coding agents vs SRE agents are different beasts
We're asking the wrong question about AI agents in production
Dogfooding is an experiment, not a mandate
Founders should code a little
Kill the flies before fighting fires
Deploy AI agents with domain-confident teams first
Engineering teams' informal failure banks
Agent prompts evolved from yelling to specs
When your agent evals get too sentient
Small anomalies compound into catastrophic outages
Cross-team debugging friction is the real killer
I quit my job to tackle complexity in software systems