Databricks’ Ali Ghodsi Never Wanted to Be CEO. Now He’s Among the Best

Summary

Ali Ghodsi, co-founder and CEO of Databricks, discusses his path to CEO, management philosophy, the Snowflake rivalry, the Lakehouse category, hiring and sales, AI adoption, and Databricks' IPO timing. The interview is mostly leadership and company-building commentary, with no explicit public-market buy or sell recommendation. Market-relevant points include Databricks waiting for calmer public markets to IPO, Snowflake's competitive weaknesses, and slow enterprise AI diffusion.

• Ali Ghodsi recounts becoming Databricks CEO on a trial basis and building an unconventional CEO playbook.
• He advocates focusing the company on one major bottleneck for one to three years.
• Databricks shifted from PLG to enterprise sales and revamped its executive team.
• The Snowflake rivalry is framed through Snowflake's proprietary stack, weak AI support, and high cost, which Databricks attacked with Lakehouse, open formats, AI, and TCO.
• Ali says AI diffusion will take at least a decade because enterprises must re-engineer processes and provide context.
• He says Databricks would be worth more public today but will wait for calmer markets; Anthropic and OpenAI have different capital needs.
• Public markets are described as fickle around AI, SaaS, and semiconductor narratives.
• The host's takeaways emphasize conflict tolerance, conviction, and patience, especially for second acts.