Lenny's PodcastTomer Cohen: How LinkedIn collapsed PM into full-stack pods
Through bespoke trust, growth, and research agents, small pods own ideas end-to-end; LinkedIn sunset its APM program and built a full-stack builder ladder.
At a glance
WHAT IT’S REALLY ABOUT
LinkedIn reboots product roles with AI-driven full stack builders
- LinkedIn’s CPO Tomer Cohen explains a radical shift from traditional, highly specialized product development to an AI-first “Full Stack Builder” (FSB) model. In response to rapidly changing job skills and rising organizational complexity, LinkedIn is collapsing processes and org structures so individual builders (and small pods) can take ideas from insight to launch, aided heavily by custom AI agents. They’ve sunset their APM program, introduced a new Associate Product Builder track, created an official Full Stack Builder title and ladder, and are re-architecting their platforms so AI can reason over LinkedIn’s systems. Cohen stresses that tools alone aren’t enough: cultural change, incentives, and deliberate change management are critical for companies that want to match the pace of technological change.
IDEAS WORTH REMEMBERING
5 ideasCollapse process and organizational complexity around empowered builders.
LinkedIn observed that product work itself isn’t inherently complex, but years of sub-steps, reviews, and micro-specialization slowed everything down; the FSB model pulls work back into fewer hands so builders (and small pods) can own problems end-to-end.
Redefine what humans uniquely own: vision, empathy, communication, creativity, judgment.
Cohen’s goal is to automate as much of the mechanical work as possible and focus human builders on high-leverage traits—especially judgment in ambiguous situations—while AI handles research, analysis, maintenance, and much of the execution.
Invest heavily in AI-ready platforms and custom agents, not just off-the-shelf tools.
LinkedIn is re-architecting UI and backend systems so AI can reason over them, and has built bespoke agents (trust, growth, research, analyst, coding, maintenance, QA) tuned to LinkedIn’s data, patterns, and standards; they found generic tools rarely work “out of the box.”
Curate and structure your knowledge base; don’t just expose everything to AI.
Naively giving agents access to all internal docs and code led to hallucinations and poor prioritization; LinkedIn now builds curated “golden” datasets and carefully designed context windows so agents can provide reliable, high-quality guidance.
Use pilots, pods, and internal success stories to drive cultural adoption.
LinkedIn seeded FSB practices in select pods and leadership teams, celebrated wins (e.g., researchers becoming growth PMs, BD leaders shipping portals themselves), and used these stories to create demand, FOMO, and momentum before wider rollout.
WORDS WORTH SAVING
5 quotesWhen we look at the skills required to do your job, by 2030 it will change by 70%. So whether or not you're looking to change your job, your job is changing.
— Tomer Cohen
The goal is to empower great builders to take their idea and to take it to market, regardless of their role in the stack and which team they're on.
— Tomer Cohen
Everything else I'm working really hard to automate.
— Tomer Cohen
It's not enough to give them the tools. You have to build the incentives programs, the motivation, the examples to how you do it.
— Tomer Cohen
If you're looking for a formal reorg or declaration to start building differently, you are waiting too long.
— Tomer Cohen
High quality AI-generated summary created from speaker-labeled transcript.
Get more out of YouTube videos.
High quality summaries for YouTube videos. Accurate transcripts to search & find moments. Powered by ChatGPT & Claude AI.
Add to Chrome