Dan Shipper on How to Ship AI Lab Hacks to Core Products
Dan Shipper explains how Every uses a staged research pipeline to turn disposable AI prototypes into core product features.
How to build products on a moving frontier | Dan Shipper (Every)
With Dan Shipper · Friday, September 25, 2026
Every co-founder and CEO Dan Shipper explains how product teams can innovate on rapidly advancing AI frontiers by separating divergent lab experimentation from core product execution. He outlines how small, two-person teams pairing 'pirates' and 'architects' can rapidly test ideas against real work and filter the best prototypes through a staged research pipeline into the main product.
Key takeaways
- Every CEO Dan Shipper separates exploratory AI lab experiments from core roadmap execution to prevent experimental chaos from breaking stable releases. Read more →
- Dan Shipper expects an applied AI lab to throw out roughly 90% of what it builds. Read more →
- Dan Shipper and the team at Every built "Kate Bench" by feeding three years of historical edits from Editor-in-Chief Kate Lee into frontier AI models. Read more →
- Dan Shipper argues that forcing engineers to explore new AI models while executing sprints ruins both workflows. Read more →
- Jeff Bezos designed the two-pizza team (8 to 10 people) for an era before modern AI tools collapsed execution time. Read more →
5 articles from this episode
Dan Shipper: How to Make Failed AI Prototypes Pay for Themselves
Dan Shipper shares how Every tests AI experiments on internal work and turns failed prototypes into marketing content and early adopter fuel.
How Dan Shipper Automated Editorial Taste with Kate Bench
Dan Shipper explains how Every built Kate Bench from three years of editorial edits, cutting manual copy-editing time by 12 percent.
Why Dan Shipper Splits Product Teams into Labs and Execution
Dan Shipper explains why product orgs need separate labs and execution teams to survive rapid AI model shifts.
Dan Shipper: Why 2-Slice Teams Beat 2-Pizza Teams in AI
Dan Shipper explains why AI product prototyping requires two-slice teams of pirates and architects instead of traditional 10-person squads.
More Lenny's Podcast episodes
- Why Google wants to win the frontier conversation | Tulsee Doshi (Head of Product for Gemini)Nov 8 · 3 articles
- 86 minutes of paid-UGC alpha | Hector Serrano (Wispr Flow)Nov 1 · 5 articles
- An 86-minute guide to finding PMF | Rob Snyder (Author, The Power of Pull)Oct 25 · 4 articles
- A mind-expanding conversation on how models actually work | Emmanuel Ameisen (Anthropic)Oct 18 · 6 articles