Michael Affronti is the Chief Product and Business Officer at DriveCentric, the number one CRM and engagement platform for automotive dealerships, and one of the most hands-on product leaders building and using AI agents today. He spent three and a half months earlier this year back in the code, learning how agents actually work from the bottom up, memory, state management, harnesses, before returning to the executive seat. In this episode of Product Talk, Sonjoy Ganguly sits down with Michael to unpack how he runs a personal fleet of over a dozen agents (named Jarvis at home, Q at work), how his AI Labs team ships two production agents a month, and what product leadership actually looks like when the team is a mix of people and agents.
What you’ll learn:
1. Why Michael says the most important thing AI does for product managers isn’t automation, it’s shortening the distance to a decision
2. How he built a personal agent fleet for everything from Sunday meal planning to conference talk prep, and what happened when agents went rogue
3. The SDLC next-gen project at Drive Centric: mapping every step from idea to shipped product, then cutting everything that doesn’t require human judgment
4. Why product briefs are actually getting longer at Drive Centric while everything else in the process is getting shorter
5. How the AI Labs team operates as single-person full-stack companies, owning the brief, the code, the customer conversation, and the launch
6. Why being too restrictive with AI access is the biggest mistake companies are making right now
7. What the best product leaders will look like in five years, and why portfolio thinking is the skill that won’t get automated away
In this episode, we cover:
(0:51) Michael’s career arc from Microsoft Outlook to Salesforce Payments to Bumble trust and safety to Drive Centric
(8:44) How the role of product management shifted from tactical launches to business impact ownership
(10:52) How AI changed his leadership style, and what he’s deliberately kept the same
(11:02) Shortening the distance to a decision: prototypes, voice notes, and same-day customer feedback
(13:29) Why one-on-ones are becoming less frequent but more meaningful
(14:17) The “back to being builders” thesis and AI’s shift toward judgment over artifacts
(16:41) The SDLC next-gen project: mapping every step from idea to shipped product and cutting the rest
(19:41) Why engineers are now joining customer meetings, and what that signals
(20:53) The personal agent fleet: Jarvis at home, Q at work, and what went wrong
(24:41) How Drive Centric built agent-building into company OKRs
(26:01) Where companies are making mistakes with AI, and the AI Labs experiment
(28:26) What the best product leaders will look like in five years
Blog and detailed workflow walkthroughs from this episode: [link]
