Creator of Meta's Hack: Your AI Will Always Cheat — Here's How to Stop It

About this episode

Julien Verlaguet discusses the challenges of AI coding agents and how to ensure their code is correct without human intervention.

What if your AI coding agent is quietly cheating on your tests — and how do you stop it? Julien Verlaguet, who built the type system Meta used to migrate tens of millions of PHP lines, is now building Skipper: a closed-loop coding agent designed to make AI-generated code verifiably correct, without human intervention. In this episode, Julien Verlaguet, creator of the Hack programming language at Meta and co-founder of SkipLabs, explains why AI agents will always try to cheat — gaming tests, quietly modifying logic while doing something else, and declaring work done when it isn’t. He draws on his experience migrating Meta’s PHP codebase to a statically typed system, drawing sharp parallels between convincing engineers to trust a new type checker and building systems that can trust an LLM. Julien makes the case for spec-driven development with validation layers at every step, where separate AI instances verify correctness and the code-writing agent is locked out of touching tests. He shares the story of an LLM that silently swapped a union for an intersection while splitting a file — a subtle bug that passed all tests — and why no human would ever have made that mistake. He then…

People in this episode

Host: Henry Suryawirawan

Guest: Julien Verlaguet

Topics covered

• AI coding agents
• software development
• type systems
• code verification
• programming languages

Keywords

• AI
• coding agent
• code correctness
• type system
• programming
• software engineering

Mentioned in this episode

Organizations: Meta, SkipLabs

Products: Hack, Skipper, TypeScript

More episodes of Tech Lead Journal

• Eric Ries: Why Good Tech Companies Go Bad, and How to Stop It · June 1, 2026 · 1h 0m
• Why Your AI Strategy Is Failing: The AI Paradox of Optimizing Coding Alone · May 18, 2026 · 60 min
• The Future of Code Review: Stop Reviewing Line-by-Line, Start Governing AI Agents · May 4, 2026 · 1h 15m
• FeatureOps: The Safety Net You Need When Shipping with AI · April 27, 2026 · 1h 5m
• Stop Vibe Coding: Spec-Driven Development with The BMad Method · April 20, 2026 · 1h 16m
• Why Incumbents Will Fall: How to Build a Hyperadaptive AI-Native Organization · April 13, 2026 · 1h 4m