Tech Lead Journal

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 walks through how Skipper works: you write a spec, hand over control, and a compiler-like agent produces correct, runnable TypeScript without back-and-forth, backed by a sound incremental type system, reachability analysis, and a reactive runtime that applies diffs in milliseconds.
He closes with a grounded take on how the developer role is shifting — not disappearing — toward the kind of design, integration, and oversight work that always mattered most.
Key topics discussed:
Why AI agents will always try to cheat on your tests
The union-vs-intersection bug an LLM introduced silently
Spec-driven development to keep LLMs on track
How to separate the AI that verifies from the one that fixes
Skipper: a compiler-like closed-loop coding agent
Sound, incremental TypeScript built for AI-speed iteration
Hot-reloading state without restarting — in milliseconds
Why developers are all becoming tech leads
Timestamps:
(00:00:00) Trailer & Intro
(00:02:34) How Did Julien Create the Hack Programming Language at Facebook?
(00:05:53) Does Static Typing Make Your Code More Secure?
(00:09:54) How Did You Convince Facebook Engineers to Adopt Hack at Scale?
(00:17:15) How Can Engineers Overcome Skepticism Toward AI Coding Tools?
(00:22:44) Should Junior Engineers Trust AI-Generated Code?
(00:29:44) How Do You Build Reliable Guardrails for LLM-Generated Code?
(00:42:15) What Validation Strategies Prevent AI Agents From Cheating on Tests?
(00:45:54) What Is Skipper and How Does a Closed-Loop Coding Agent Work?
(00:54:59) How Does Skipper Compare to Claude Code in Terms of Correctness?
(00:58:27) How Do You Get Started With Skipper and What Does the Output Look Like?
(01:04:50) How Will the Software Developer Role Change in an AI-First World?
(01:09:06) 3 Tech Lead Wisdom
_____
Julien Verlaguet’s Bio
Julien Verlaguet is a programming language designer and the Founder and CEO of SkipLabs. He is best known as the creator of Hack—the gradually typed language he built at Facebook that currently powers over 100 million lines of the company’s production code. After creating the open-source reactive framework Skip, Julien founded SkipLabs in 2022. His company recently launched Skipper, a closed-loop coding agent that takes a single prompt from a developer and returns a running, validated service.
Follow Julien:
LinkedIn – linkedin.com/in/julien-verlaguet-b5710a20
X – x.com/JulienVerlaguet
SkipLabs - skiplabs.io
Skipper - skipperai.dev
Skipper’s Discord – discord.gg/bsnXyw2F9P

Like this episode?
Show notes & transcript: techleadjournal.dev/episodes/260.
Follow @techleadjournal on LinkedIn, Twitter, and Instagram.
Buy me a coffee or become a patron.

Why do companies with the best intentions end up betraying their customers, employees, and mission? Eric Ries calls it “financial gravity” — an invisible force that pulls even the most principled companies toward corruption, and understanding it is the first step to resisting it.
In this episode, Eric Ries, entrepreneur and author of The Lean Startup and Incorruptible, shares why building a great company isn’t just about having a strong vision — it’s about building structures that protect that vision from external pressure. Eric revisits the core ideas behind the Lean Startup and MVP, explaining how the purpose of a minimum viable product is not to ship fast but to learn fast. He then introduces the central thesis of his new book: that the corruption we see in companies isn’t caused by bad people, but by a financial system that pulls organizations away from their values. Drawing on stories of Sol Price, FedMart, Costco, HEB, Novo Nordisk, and Anthropic, he shows that incorruptible companies are built through a combination of ethos — a deep operational commitment to doing right — and structural governance that resists outside pressure. He also unpacks how false metrics like OKRs can hollow out a company’s integrity over time, and how Mary Parker Follett’s concept of the “invisible leader” helps culture survive beyond any single founder or CEO.
Key topics discussed:
What “financial gravity” is and why even good companies fall to it
The true purpose of an MVP (hint: it’s not about shipping fast)
Why OKRs become dangerous false proxies over time
Blueprint for building a truly incorruptible company
Why Costco and Novo Nordisk resisted forces that killed FedMart
Mary Parker Follett’s invisible leader explained
Why Anthropic’s structure gives it a lasting competitive edge
How everyday decisions become acts of systemic change
Timestamps:
(00:00) Trailer & Intro
(02:31) What Two Mega-Trends Make Lean Startup More Relevant Than Ever?
(04:03) What Is the True Purpose of a Minimum Viable Product?
(11:04) Has AI Actually Made Building Software Cheaper and Better?
(13:41) What Two Stories Inspired the Book Incorruptible?
(20:38) What Is Financial Gravity and Why Does It Corrupt Even Good Companies?
(26:29) What Is Surrogation and Why Do OKRs Become Dangerous False Proxies?
(29:55) What Is the Blueprint for Building an Incorruptible Company?
(33:53) What Is the Invisible Leader and How Does It Keep Company Culture Alive?
(39:56) What Governance Structures Can Shield a Company’s Mission from Financial Gravity?
(48:27) Why Does Anthropic’s Unique Structure Give It a Competitive Advantage in AI?
(51:43) 3 Tech Lead Wisdom
_____
Eric Ries’s Bio
Over the last two decades, Eric Ries’s ideas about continuous innovation, long-term thinking, governance, and market reform have reshaped company building and management practices. He is the creator of the Lean Startup method, and the author of the New York Times bestseller The Lean Startup; The Leader’s Guide; and The Startup Way.
As a founder, he has put his own ideas into practice with The Long-Term Stock Exchange (LTSE); Answer.AI, an AI R&D lab; Virgil, a legal services startup; and IMVU. On The Eric Ries Show, he talks with world-class technologists, thought leaders, and executives building for the long-term. He lives in the San Francisco Bay Area with his wife and three children.
Follow Eric:
LinkedIn – linkedin.com/in/eries
X – x.com/ericries
Podcast – www.ericriesshow.com
Website – incorruptible.co
Newsletter – news.theleanstartup.com

Like this episode?
Show notes & transcript: techleadjournal.dev/episodes/259.
Follow @techleadjournal on LinkedIn, Twitter, and Instagram.
Buy me a coffee or become a patron.

What if faster coding is actually slowing your software delivery down? Most teams are pouring AI into the coding phase, but the real bottleneck is everywhere else.
In this episode, Andrew Haschka, Field CTO at GitLab for Asia Pacific and Japan, explains why most AI strategies in software engineering are failing and what it takes to fix them. He introduces the AI paradox: teams invest heavily in AI-assisted coding, yet coding accounts for less than 20% of the software delivery lifecycle, leaving the biggest bottlenecks untouched.
Andrew makes the case for intelligent orchestration — moving from isolated AI interactions to governed, end-to-end agentic flows that span planning, coding, testing, security, compliance, and release. He shares how a unified system of record forms the foundation for high-quality AI outcomes, and why fragmented tools and siloed context actively limit what AI can deliver. Drawing on real customer examples — including Ericsson’s 50% faster deployments and 130,000 hours saved in six months — he shows what a holistic approach actually looks like in practice.
The conversation also covers how tech leads, developers, and junior engineers need to evolve their skills in a world where AI handles routine implementation. Andrew closes with a compelling argument: in the agentic era, governance isn’t just a compliance burden, it’s the primary source of competitive advantage.
Timestamps:
(02:30) What Are the Key Responsibilities of a Field CTO at GitLab?
(03:26) Why Should Organizations Govern AI Strategy Rather Than Chase the Latest Features?
(06:41) Why Is an End-to-End Agentic Flow More Valuable Than Individual AI Tools?
(09:39) What Is the AI Paradox and How Does Intelligent Orchestration Solve It?
(14:47) How Does Shifting Focus to Requirements Quality Transform Software Delivery Outcomes?
(18:19) How Has GitLab Evolved Beyond CI/CD Into a Full End-to-End Delivery Platform?
(20:20) What Should Software Teams Prioritize Beyond Coding in the AI Era?
(24:14) How Do Organizational Silos Create a Capability Threshold for AI Adoption?
(27:49) What Practical Strategies Can Organizations Use to Break Down Internal Silos?
(30:58) How Did Ericsson Achieve 50% Faster Deployments and Save 130,000 Hours With GitLab?
(33:07) How Should Software Developers Evolve in the Age of AI Agents?
(36:26) How Is the Tech Lead Role Evolving in a Hybrid Human-AI Team?
(39:22) How Can Junior Developers Keep Up With the Rapid Shift in Industry Expectations?
(42:40) Why Do 79% of Singapore DevSecOps Practitioners Believe AI Will Create More Jobs?
(45:27) Why Are Companies Reducing Staff Despite the Growing Demand for Software?
(48:34) What Are the Most Common Pitfalls When Implementing Agentic Workflows?
(52:29) What Practical Steps Should Engineering Leaders Take to Govern AI Responsibly?
(55:13) Why Should Engineering Leaders Build an AI Strategy Before Choosing Technology?
(57:15) 3 Tech Lead Wisdom
_____
Andrew Haschka’s Bio
Andrew Haschka serves as Field CTO for Asia Pacific & Japan at GitLab, where he acts as a trusted strategic advisor to enterprise customers and partners navigating complex technology transformation. With over 20 years of experience spanning software delivery, cybersecurity, cloud infrastructure, and organisational transformation, Andrew brings a rare combination of technical depth and executive-level counsel to the organisations he works with.
Prior to GitLab, Andrew held senior leadership roles across APAC at Google and VMware, and has led large-scale digital transformation programmes for organisations including Downer, IBM, Jones Lang LaSalle, Thomson Reuters, Optus, and across the Fiji and Pacific Islands.
Follow Andrew:
LinkedIn – linkedin.com/in/andrewhaschka

Like this episode?
Show notes & transcript: techleadjournal.dev/episodes/258.
Follow @techleadjournal on LinkedIn, Twitter, and Instagram.
Buy me a coffee or become a patron.

(07:22) Brought to you by Mailtrap
Mailtrap is a modern email delivery for developers with native SDKs support along with security compliant API & SMTP. Plus, you get 4,000 emails a month completely on their free tier! It also provides 24/7 support where you actually talk to real people, not an AI chatbot. Try Mailtrap for free at ⁠mailtrap.io⁠.

What does code review mean when AI writes most of the code? The answer isn’t to review more carefully. It’s a fundamentally different process, one built around rules, agents, and governance rather than diffs and comments.
In this episode, Itamar Friedman, founder and CEO of Qodo.ai, shares how AI is forcing a complete rethink of code review — from inline comments on code diffs to multi-agent governance systems that verify intent, architecture, and business logic at scale. He traces the evolution of code review through successive generations, explains why traditional static analysis is no longer sufficient, and lays out what a modern quality and governance layer actually looks like. Itamar also introduces the concept of “shift up” — extending quality checks into the planning phase so that technical product managers can contribute directly to shipping features — and explains how teams can move from vibe coding to viable, grounded development. The conversation also covers the race between AI labs, the role of open-source models, and a frank look at where the software developer role is heading by 2030.
Key topics discussed:
Why line-by-line code review doesn’t scale with AI-generated PRs
The generational evolution of code review tools (Gen 1 to 3.5)
How multi-agent systems surface only what needs human attention
Turning tribal knowledge into enforceable rules and skills
Shift-left and shift-up: embedding quality earlier in the workflow
What the new agentic code review UI will look like
Vibe coding vs. viable coding: the governance layer in between
Where the software developer role is headed by 2030
Timestamps:
(00:00:00) Trailer & Intro
(00:02:50) How Has AI Driven the Evolution of Code Review to Multi-Agent Systems?
(00:07:53) How Do We Move from Vibe Coding to Viable, Grounded Development?
(00:12:35) Are Traditional Static Analysis Checks Still Sufficient in the AI Era?
(00:16:27) How Do We Handle Exploding PR Volume Without Sacrificing Code Review Quality?
(00:22:11) How Do We Evolve Code Review from Simple Comments to Senior-Level AI Reviews?
(00:28:51) What Will the New Agentic Code Review UI Look Like?
(00:33:32) How Does Qodo Differentiate Itself as an AI Code Review and Governance Platform?
(00:37:15) What Do Shift-Left and Shift-Up Mean for the Future of Code Quality?
(00:41:23) How Do We Maintain Quality When Running Multiple AI Agents in Parallel?
(00:48:11) How Are Chinese AI Models Reshaping the Open-Source vs Closed-Source Race?
(00:55:25) Which AI Models Excel at Code Review, and Are We Heading Toward Specialization?
(01:03:16) Will Software Developers Still Be Needed as AI Automates More of Engineering?
(01:08:50) 3 Tech Lead Wisdom
_____
Itamar Friedman’s Bio
Itamar Friedman is the CEO and Co-Founder of Qodo, an AI code review platform used by 1M + developers. Before founding Qodo, Itamar was a founder of Visualead, which was acquired by the Alibaba Group. He then worked for Alibaba Group for 4 years as the Director of Machine Vision. Now, Itamar is dedicated to quality-first code generation.
Follow Itamar:
LinkedIn – linkedin.com/in/itamarf
X (formerly Twitter) – @itamar_mar
Qodo.ai – qodo.ai

Like this episode?
Show notes & transcript: techleadjournal.dev/episodes/257.
Follow @techleadjournal on LinkedIn, Twitter, and Instagram.
Buy me a coffee or become a patron.

(05:00) Brought to you by Mailtrap
Mailtrap is a modern email delivery for developers with native SDKs support along with security compliant API & SMTP. Plus, you get 4,000 emails a month completely on their free tier! It also provides 24/7 support where you actually talk to real people, not an AI chatbot. Try Mailtrap for free at mailtrap.io.

What happens when AI ships code faster than your team can review it? As agentic development accelerates your SDLC, the guardrails matter more than ever — and most teams don’t have them.
In this episode, Egil Osthus, CEO of Unleash, makes the case for FeatureOps as a strategic capability — not just a developer convenience. He explains the shift from a project mindset to a product mindset, where releases are decoupled from deployments and business outcomes matter more than shipping scope. Egil breaks down the four pillars of FeatureOps — gradual rollout, full stack experimentation, surgical rollback, and lifecycle management — and why each one becomes even more critical as AI-generated code flows faster into production. He also warns against building your own feature flag solution in-house, and shares what the rise of agentic development means for engineers who must now act as guardians of an oversight layer.
Key topics discussed:
Project mindset vs. product mindset in software delivery
The 4 pillars of FeatureOps and what each one solves
Why feature flags scare executives — and how to win them over
Decoupling deployment from release across Dev, PM, and Marketing
The danger of rolling your own feature flag solution
How local evaluation keeps feature flags fast and private
Blast radius management in an AI-accelerated SDLC
What vibe coders get wrong about day-two operations
Timestamps:
(00:00) Trailer & Intro
(02:36) What Is the Current State of Feature Flag Adoption Across the Industry?
(05:32) Why Is Feature Flag Adoption So Challenging Despite Its Apparent Simplicity?
(10:44) How Does FeatureOps Differ From CI/CD and Progressive Delivery?
(12:26) What Are the Four Core Pillars of FeatureOps?
(16:11) How Can Teams Shift the Perception of Feature Flags From Tactical to Strategic?
(20:46) How Do Feature Flags Align the Needs of Developers, Product Managers, and Marketing?
(25:09) How Do Organizations Effectively Define Responsibilities for Strategic Feature Flags?
(28:03) Does Using Feature Flags Enable Your Team to Deploy on Fridays?
(30:41) What Is Unleash and How Does It Scale for Enterprise Needs?
(34:54) What Are the Hidden Dangers of Building Your Own Feature Flag Solution?
(39:32) Why Are Local Evaluation and Privacy Core to Unleash’s Design?
(44:48) How Does the Rise of AI Impact the Evolution of FeatureOps?
(52:02) What Specific Guardrails Does FeatureOps Provide to Improve Safety?
(54:21) Can FeatureOps Platforms Use AI to Autonomously Manage Feature Rollouts?
(55:33) What Essential FeatureOps Advice Should Every Vibe Coder Follow?
(59:53) 3 Tech Lead Wisdom
_____
Egil Osthus’s Bio
Egil Østhus is the co-founder and CEO of Unleash, the world’s leading open-source feature management platform. As a seasoned enterprise technologist and product strategist, he operates at the cutting edge of business and software engineering.
Egil’s mission is to help technology leaders and businesses move beyond traditional DevOps by embracing FeatureOps, a new methodology that provides a critical safety net for the accelerating, and often risky, world of agentic software development. He has a unique ability to speak the language of both engineers and senior executives, making complex topics accessible and actionable.
Follow Egil:
LinkedIn – linkedin.com/in/egilconr
Unleash – getunleash.io

Like this episode?
Show notes & transcript: techleadjournal.dev/episodes/256.
Follow @techleadjournal on LinkedIn, Twitter, and Instagram.
Buy me a coffee or become a patron.