Databricks' decade-long MLOps buildout vs. Snowflake's AI/ML lag, in one thread
hu_yifei · x · 2026-10-29
huyifei traces Databricks' MLOps/MLSys accumulation: MLflow (2018, MLOps for classic ML), Cortex (2022, MLOps for deep learning), and Mosaic (2023). Snowflake, by contrast, has always trailed in AI/ML systems, staying strong mainly in BI friendliness for non-technical users.
Hanchung Lee adds that Mosaic did substantial image-model training optimization before ChatGPT, with an early Oracle Cloud GPU collaboration in late 2022/early 2023.
More from Companies & People
- Anthropic spent $2.75M guarding Dario Amodei—over 4x his base salary — FinanceYF5 · 2026-10-30
- Healthcare company's ops team burns 1.2B Opus tokens in a single day — RishabJainK · 2026-10-30
- Cloudflare developer users grow 10x in 3 years, from 1M to 10M — yihui_indie · 2026-10-30
- AMD and ai& to host one-day Tokyo hackathon: 8 teams, 4.5 hours, 750K JPY tokens per winner — DavidBennett__ · 2026-10-30
- Advice for lost CS majors: add AI agents to your favorite game and publish results — Aizkmusic · 2026-10-30
- Sam Altman Reveals Peter Thiel's Advice That Made OpenAI Double Down on ChatGPT — davidsenra · 2026-10-30