Ali Ghodsi on becoming Databricks' CEO and the bottleneck playbook

Key points

Key takeaways from a Sequoia Capital interview with Databricks CEO Ali Ghodsi (September 2026):

Transcript unavailable — summary based on title and description only.

CEO as a trial run. In 2015 Databricks had a successful open source project and $1.5M in revenue, but the board was quietly interviewing other candidates behind Ghodsi's back; he got the seat as a trial run, without a CEO salary.

The compliment. Ben Horowitz told the interviewer that he considers Ghodsi the best CEO in Silicon Valley outside his own portfolio.

The bottleneck playbook. Ghodsi finds the single biggest bottleneck in the company, then applies an absurd, disproportionate amount of attention to it for two to three years — how the Lakehouse category got built while consultants, sellers and most of the internet dismissed it.

Fighting Snowflake. He breaks down the playbook Databricks ran against Snowflake category-creation, positioning and enterprise motion.

Saying no. Ghodsi turned down phenomenal executives who still hold a grudge, explaining his filters for leadership hires.

Lessons learned. The interview covers what a decade of founder-CEO leadership taught him about focus, sequencing and what actually moves a data company.

Read more: Sequoia Capital — Databricks' Ali Ghodsi Never Wanted to Be CEO