First a standard recruiter screen covering my background and ownership experience, followed by a 60-min phone screen with TPM split into three parts: program sense, technical depth, and behavioral. Was asked about balancing competing priorities between engineering and product, plus a technical bit on designing a global service with failover, and also brushing up on using AI effectively. The onsite was a five-round back to back. One was pure TPM execution, focusing on driving large programs, dependencies, and OKRs. Then there was a sys design round where I had to walk through scalable architecture, latency, and consistency tradeoffs. Another round was product sense, which was more about translating cross-functional goals into actual deliverables. The Googleyness round was all about conflict resolution and influencing without authority, and the final was engineering collaboration, where I had to scope tradeoffs with infra and AI teams. When you prep, make sure you can go a level deeper on the technical pieces instead of staying high level. I realized pretty quickly that hand-wavy answers don’t work in those rounds. I did a few mock interviews with Google TPMs and that honestly helped me manage my time better and not ramble during the 60-minute interviews.