Ai on Quentin Rousseau

Your Small PR Rule Won't Survive AI

Tue, 19 May 2026 08:00:00 +0000

For two years, we enforced a strict small-PR culture at Rootly . Stacked PRs, atomic changes, never more than a couple hundred lines. The logic was sound: smaller diffs are easier to review, easier to revert, easier to reason about.

Then AI started writing most of our code.

Two Mornings a Day

Wed, 08 Apr 2026 08:00:00 +0000

I haven’t slept a normal night in over four years. Somehow, I feel better than I used to.

When I started Rootly , my sleep broke. US customers needed me in the morning. Features needed shipping at night. My brain refused to stop planning in between. I wasn’t grinding on purpose. I was building a company across time zones, and my body just… adjusted.

Watts, Water, and Carbon: What Your AI Prompts Actually Cost

Thu, 19 Mar 2026 12:00:00 +0000

Every time you ask an AI to summarize a document, generate an image, or write code, a GPU somewhere lights up — and a power meter ticks upward. As someone who runs AI agents all day, I started wondering: what’s the actual energy footprint of this revolution I’m participating in?

The answer is more nuanced than the headlines suggest.

What a single prompt actually costs

Let’s start small. You’ve probably seen the stat that a ChatGPT query uses “10x more energy than a Google search.” The reality is more complicated . Google’s traditional search uses about 0.3 watt-hours. The Electric Power Research Institute estimated a ChatGPT request at 2.9 Wh — roughly 10x.

One More Prompt: The Dopamine Trap of Agentic Coding

Mon, 09 Mar 2026 14:00:00 +0000

It’s 2:47 AM on a Tuesday. I’m not debugging a production outage. There’s no deadline. I’m just watching Claude Code refactor a module — and I can’t stop. One more prompt. One more agent run. One more hit.

If this sounds familiar, you’re not alone. Something strange is happening to developers in 2026: AI coding tools may be triggering the same dopamine loops as slot machines, and the tech industry is sleepwalking into it.

From Writing Code to Orchestrating Agents

Mon, 09 Mar 2026 08:47:29 +0000

My dev workflow has changed more in early 2026 than in the previous five years. I went from writing code in VS Code, to pair-programming with AI in Cursor, to running CLI agents like Claude Code and Codex — and now I’m orchestrating multiple agents working in parallel on different tasks at the same time.

Each step felt like a leap. But the jump to multi-agent orchestration is the one that fundamentally changed how I think about shipping software.

Your AI Agent Doesn’t Need an MCP Server — It Needs a Good CLI

Tue, 24 Feb 2026 23:27:05 +0000

Every week, a new MCP server shows up on GitHub that wraps a CLI tool that already existed. MCP server for GitHub? gh has been doing that since 2020. MCP server for AWS? The aws CLI covers 200+ services with --output json. MCP server for Datadog? They shipped a CLI instead — and it works better.

I get the excitement around MCP (Model Context Protocol). It’s a clean standard for connecting AI to external systems. Anthropic designed it well. But somewhere along the way, the community started treating MCP as the default way to give AI agents access to tools — even when a perfectly good CLI already exists.

From "Maintainers Wanted" to Three Major Releases in a Weekend

Tue, 24 Feb 2026 22:42:22 +0000

I have 50+ public repositories on GitHub. Some of them haven’t been touched in years. A few still get stars and forks weekly. Most sit somewhere in between — not abandoned, not active, just… there.

Until recently, that “just there” state came with guilt. Every Dependabot PR I didn’t merge, every issue I didn’t respond to, every README that still referenced a deprecated API — it all felt like debt accumulating. Maintaining open-source repositories was a second job I never signed up for.

Rethinking How We Evaluate Product Engineers in the AI Era

Wed, 18 Feb 2026 12:24:08 +0000

Over the past few years as CTO at Rootly , I’ve had to fundamentally rethink how we evaluate product engineers. The skills that used to separate great candidates from good ones have shifted — not because the bar is lower, but because AI has changed what “doing the work” actually looks like.

Back Online: From Octopress to Hugo, From Silence to AI

Tue, 17 Feb 2026 23:32:48 +0000

My last blog post was in February 2017. Nine years ago.

It wasn’t intentional. I didn’t wake up one day and decide to stop writing. I just… didn’t start the next post. Then a month passed, then a year, then almost a decade. The classic // TODO: write blog post that never gets resolved.

What happened

Life happened — and it was a busy one.

When I wrote that last post about PgBouncerHero in 2017, I was at Instacart as one of their early SREs, building the infrastructure that took the platform from thousands of orders a week to millions.