How BabyList Accelerated AI Adoption in Engineering with Karynn Ikeda

12 snips

Mar 20, 2026

Karynn Ikeda, former engineering manager turned AI enablement lead who guided BabyList’s rollout of agentic IDEs and Cloud Code. She recounts early tool pilots, a six-week Windsurf test, company-wide agent experiments, onboarding PMs and designers into code, and the shift toward orchestrating multiple coding agents while measuring developer sentiment over raw velocity.

Ask episode

AI Snips

Chapters

Transcript

Episode notes

INSIGHT

Cloud Code Consolidation Followed Organic Engineer Preference

Adoption moved from agentic IDEs to cloud-based Claude Code, then the company consolidated and invested in Cloud Code skills/plugins.
Engineers organically preferred cloud models, making vendor lock-in less of a concern and enabling centralized skills.

ADVICE

Get Leadership Budget And Staff Engineers Driving Decisions

Do secure leadership buy-in and a staff-level working group to speed procurement and craft best practices for AI tools.
BabyList gave the CTO a generous budget and had staff engineers interview teams and spike on memory/context solutions.

ADVICE

Measure AI By Developer Sentiment Not Just Velocity

Do measure AI adoption by developer sentiment rather than only raw velocity metrics to capture real usefulness and confidence.
Use existing sentiment tools or simple weekly surveys and iterate rather than over-engineering complex measurements.

Get the Snipd Podcast app to discover more snips from this episode