#engineering

With Thomas Pierrain and Vanessa Gossent, I revisit AI costs, hiring, parallel agents and the work we have to redo after delivery.

My conversation with Bruno Soulez on context engineering, agent harnesses and code review: what I delegate and what I still check.

Three months, 1,011 pull requests and 8,728 inline AI review comments: 4.5% got a visible human response. What the data shows, and what it cannot.

When the same accepted correction returns, test whether it belongs in a linter, type, test, example, repository rule, checklist, or architecture decision.

Compare independent review with self-review and deterministic checks. Measure rescued defects, false acceptances, regressions and the cost of another pass.

A review signal matters only when its provider, commit SHA, coverage, execution state, publication, and place in branch policy are explicit.

Build a bounded review loop with frozen inputs, separate execution and outcome fields, retry limits, evidence handoffs and human escalation.

A verified $214 bill for a production rewrite, checked against the source transcript, sits beside a corpus of about 2,900 tech talks on coding-agent costs.

Opus 5.5 changes input, output and cache economics. Price the accepted task, including retries and review, before choosing a cheaper model.

Claims that nobody can measure AI ROI and that 95% of pilots return zero are weaker than their usual retellings, once their sources are checked.

A $589 billion single-day stock rout, an enterprise model platform, and a €25,000 backup server show three scales of vendor dependence.

Personal CLAUDE.md to team AI instruction system for six engineers. How Méthode Aristote separates sources, shares modules, and catches behavioral drift in CI.

Use L0 to L5 to diagnose context drift, then maintain adherence through observation, repair, and replay instead of treating setup as finished.

Anthropic measured merged PRs; METR measured task duration in a different setting. What these results and long-context research establish, and what remains a hypothesis.

6 human commit authors plus Claude in our git history (May 2026 snapshot). What the commit patterns actually look like after months of Claude Code, beyond the marketing claims.

The concepts I wish I’d known before week one: the agent loop, instruction scopes, context management, skill invocation, hooks, and client permission checks.

Choose between Claude Code and Cowork, then follow the Start, Build, or Scale path that matches your role and current operating problem.

Nine months of AI configuration in one production project: evolving responsibilities, profile-based generation and a measured reduction in always-on lines.

Everyone talks about shipping 10x faster with AI. Nobody talks about accumulating debt 10x faster. 7 months of production data from a real EdTech platform.

A non-developer modified 80 files in production in 10 days with Cursor AI. Exact timeline, AI config setup, and what it means for engineering teams in 2026.

Jeremy Twei coined it. Addy Osmani popularized it. Margaret-Anne Storey extended it to teams. Here's what I built to fight all three.

What roughly 15 years in software, including 8 to 10 in leadership, give me as an AI-assisted builder, what I had to relearn, and why faster code is not enough.