Building products alone vs through the team

A 10-year veteran PM is hitting a wall. He built a reputation solving hard problems solo, but that playbook breaks when you become a leader of PMs. The question he's asking is brutally honest: does your team ship differently because you're there, or do they ship the same thing faster when you stay out of the way? Two things jump out. First, being the smartest person in the room stops scaling past your own effort. Second, your job as a leader is to make your team better at their jobs, not to prove you're still the best problem solver. This matters now because half the PMs reading this are either about to make this jump or already flailing through it. Knowing when to hand off beats knowing how to do everything.

AI in Product Management: Understanding the Skills and Tools Needed for the Future - Egon Zehnder

Look, AI isn't coming to product management. It's already here. This Egon Zehnder piece cuts through the hype and tells you what you actually need to stay relevant. First takeaway: technical literacy matters more than ever. You don't need to code, but you need to understand what's possible and what's not. Second takeaway: your real superpower is judgment calls. AI can generate options and analyze data faster than any human. What it can't do is decide what customers actually need and when to say no. Right now, PMs who can translate between the AI tools their team's using and the humans actually buying your product are going to be the ones running the show.

Principal AI Product Manager (Mock Interview) "How innovative are you?"

I watched this mock interview and realized most PMs are thinking about AI product decisions all wrong. The key moment was when the interviewer pushed back on "I wouldn't ship it" for a safety risk. The actual answer? You need to quantify the risk, run proper evals with your design team, and know when you're taking on technical debt versus crossing a real line. Two takeaways: First, you need a framework for safety gates before you're in the moment. Second, meta-prompting and model corrections aren't engineering problems anymore. They're product decisions. This matters now because every PM hiring for AI roles will be asked this question, and the old "safety first" platitude won't cut it anymore.

From Roadmaps to Real-Time Systems: Rethinking Product Management for the AI Agent Era

The old playbook is dying. We've spent years perfecting quarterly roadmaps and feature releases, but AI agents operate in real-time, learning and adapting faster than we can plan sprints. This piece cuts through the hype and shows what actually changes about how we prioritize, measure, and ship. The two things that hit me: first, you need to build feedback loops into your product instead of relying on quarterly cycles, and second, success metrics shift from adoption to agent effectiveness. It matters now because teams that don't adapt their PM approach to this reality will ship products that feel outdated the moment they go live.

Impostor Syndrome Among Us

Look, impostor syndrome hits PMs harder than most roles. You're constantly making decisions with incomplete information, your roadmap gets questioned by engineers and executives alike, and you're never quite sure if you're the right person for the job. Here's the thing: recognizing that feeling isn't a weakness. It's actually a sign you care about getting it right. The real takeaway is this. Stop waiting to feel 100% confident before you lead. The best PMs I know are the ones who acknowledge the doubt, ask good questions anyway, and iterate on their approach. That self-awareness keeps you grounded and your team trusting you.