Reliable by design: lower cost, better delivery, trusted AI
Guitar Center rebuilt its data stack around control, speed, and trust, cutting Snowflake spend 40% along the way.
Overview
Data teams face pressure to cut costs, ship faster, and prepare for AI at once; the teams that succeed build reliability in from the start. Guitar Center's VP of Data covers building scalable infrastructure with dbt and Airflow: predictable costs via guardrails, reduced operational friction, and trustworthy data for analytics and AI.
What you'll take away
- 01
Guitar Center rebuilt around three pillars, cost, delivery flow, and confidence in the data, after inheriting a failed migration with 2,000+ lines of stored procedures and no CI/CD.
- 02
Optimizing one hourly sales object cut Snowflake run rate 40%, and reducing Airflow minutes cut consumption another 50%, turning cost control into a continuous habit rather than a one-time fix.
- 03
Stability isn't correctness: a pipeline can run reliably every day and still be reliably wrong, and AI will
Session speaker
Mike Doll
Guitar Center