Descriptions:
Mitchell Hashimoto, co-founder of HashiCorp and creator of Terraform and the Ghostty terminal, sits down with The Pragmatic Engineer for a wide-ranging conversation that spans the origins of the HashiCorp infrastructure stack, the experience of taking the company public in 2021, and his current thinking on how AI is reshaping software engineering as a craft.
The most technically substantive sections cover Hashimoto’s daily AI coding workflow — including why he keeps an agent running continuously in the background — and his nuanced view of which tasks benefit from AI assistance and which still require focused human attention. He offers practical advice for engineers who have not yet adopted agentic tools, framing the transition not as a threat but as an expansion of what individual builders can accomplish.
The interview also tackles two bigger structural questions. First, whether open source can survive AI-generated contributions: Hashimoto argues that AI makes it trivial to produce plausible-looking but incorrect code, effectively forcing open-source communities to shift from a trust-by-default model to a trust-by-earned-reputation model. Second, whether Git itself remains relevant as AI agents increasingly author, review, and merge code — he notes this is the first time in 12–15 years that question can be asked without being dismissed. His perspective, grounded in years of building widely-adopted infrastructure tooling used by millions of engineers, gives these observations particular practical weight.
📺 Source: The Pragmatic Engineer · Published February 25, 2026
🏷️ Format: Interview
