People Who Know Things

Discussion with Aubrey Blanche, an AI ethicist, director at The Ethics Centre in Sydney, and founder of The Mathpath, about why productivity is the wrong goal for AI adoption, how to build AI policy with the people it affects, why judgment is the skill that lasts, and why joy should be a right.

Where to find Aubrey

• LinkedIn
• Website
• The Mathpath Notes (newsletter)
• People Over Platforms (newsletter)
• The bAIbes (podcast)

When every engineer can own a feature end to end with AI, who do they collaborate with? Kelsey Duncan, Senior Engineering Manager at FreeWill, shares what’s changed for her team: fewer blockers, more isolation, and open questions about what career growth looks like. We also cover moving from Cursor toward Claude Code, onboarding engineers onto AI-built repos, and why her team’s skills are named Bake and Sift.

Mike Nakhimovich led Airbnb’s Dev AI team through a no-mandate rollout that took agentic coding from 5% to 98% adoption. Now he’s building Hive, a place to co-work with your agents that feels more like a farm than a factory. We talk about where to keep humans in the loop, why async beats synchronous, and why 50–80% faster “in a happy way” beats 4X.

Hive is in closed beta as of the release of this podcast. You can join the waitlist at friendlyai.dev

Reach Mike at [email protected]

Discussion with Josh Penzell, founder of Imagination Applied, about directing AI instead of prompting it: from his work on Special Olympics’ multilingual Unified Champion Schools Inclusion Academy to why organizations should reward experimentation over time savings.

Josh's LinkedIn

Josh's Website

Special Olympics’ multilingual Unified Champion Schools Inclusion Academy

Discussion with Cora Anthony, an AI enablement and L&D leader, about getting real ROI from AI: for organizations, by starting with discovery and measuring time saved, and for individuals, by building skills that go with you wherever you work.

AI makes great engineers better, good engineers great, and bad engineers worse. Joseph Cutrono, VP of Engineering with 20+ years in software, shares how he measures AI success, why non-engineers should start the coding process, and what tokens are really costing us.

Discussion with Mac Merritt, Sr. Engineer at AWS, about combatting code slop as engineers move from using AI too little... to maybe too much.

Discussion with Mac Merritt, Sr. Engineer at AWS, about Dario Amodei's essay, "We Must Pace the Frontier." In this episode we discuss whether slowing down the pace of AI advancement is the right answer to security concerns... and is slowing down even possible?

"We Must Pace the Frontier" Essay Link

NYT Article

We discuss the recent decision to ban student-facing GenAI in New York City Public Schools with Melissa Belardi. Mel has a background in learning science, education, learning products, and has worked with AI in the corporate setting including leading several AI training programs.