Lenny's Podcast: Product | Career | Growth · Lenny Rachitsky

How AI is reshaping the product role | Oji and Ezinne Udezue

·1 hr 18 min·5 clips

Ezinne reveals how AI transforms PMs from managing static software to orchestrating living, breathing systems.

Lenny Rachitsky hosts Aji and Ezinne Udezue, a married pair of senior product leaders with over 50 years of combined experience, to discuss how AI is transforming the PM role and what skills will matter most in the years ahead. Ezinne describes the core PM value proposition as unchanged -- de-risking product delivery while maximizing business return -- but notes that the tactical execution is shifting significantly: engineering cycles are accelerating so fast that PMs risk becoming the bottleneck, and the orchestration layer has expanded to include feedback loops, LLMs, and multi-touch data flows rather than just people in rooms. Aji introduces the concept of the sharp problem as the antidote to what he calls drunken startup building: locating old, core human needs and reimagining them with new technology to produce 3-10x improvements compelling enough to generate immediate demand. He describes the shipyard model as a six-capability team -- PM, engineering, design, user research, data and ML, and product marketing -- that operates in near-continuous collaboration rather than stand-up cadence, with tendrils reaching into customer-facing functions like support managers who can flag design problems before they ship. Ezinne identifies the five attributes she screens for when hiring PMs for AI-era product work: curiosity with genuine humility to take junior AI PM courses even as a CPO; agency understood as acting like a thermostat rather than a thermometer -- changing the room temperature rather than measuring it; strong data literacy including an understanding of how data is organized and leveraged; the ability to write evals to verify LLM outputs; and comfort with multi-model orchestration, choosing between different models based on what each does best. She argues that humility is especially critical because the pace of AI change means no one can stay ahead on credentials or experience alone; the people who succeed will be those willing to keep learning from sources that predate their current seniority. Aji describes his personal hands-on learning methodology: writing more code in the past year than in the previous ten, converting PRD writing into prototype writing, calling APIs himself via Postman, and running a home automation project as a personal eval that requires him to learn quantization, fine-tuning, MCP server construction, and home assistant integration. He frames the project as a discipline: pick something you are passionate about, build it, and let the passion pull you through the learning curve. The episode also addresses the CPO versus CTPO debate, with Ezinne arguing that the tension between the CPO optimizing for opportunity and the CTO optimizing for resources is generative rather than redundant, producing sharper decisions at the strategic level. Both hosts argue that it is still day one for AI in product, that the playbooks being written now may not survive two or three years, and that the wisest approach is teachability -- which Aji equates with survivability. The tone is warm, practically oriented, and structured around concrete frameworks rather than abstract trends.

As heard by us

AI changes product work most when operators stay hands-on and keep the problem front and center.

Aji and Ezine Udezwe argue that AI does not make product work disappear; it raises the bar for judgment, practical fluency, and how well a problem is framed.

Read the full review in PlayNext →

Why you'd press play

You want a grounded read on how AI is changing product work, not just the hype.

Read the full recommendation in PlayNext →

Listen to the show on

Apple Podcasts Spotify

Play clips from this episode More from Lenny's Podcast: Product | Career | Growth