Loading...
Q3 2025 · United States
4 ai & machine learning startups based in United States ranked by GitHub engineering acceleration. Filtered from our broader AI & Machine Learning sector rankings.
| # | Company | Stage | Geo | Commits (14d) | Change | Contributors | Contrib. Growth | New Repos | Signal |
|---|---|---|---|---|---|---|---|---|---|
| 1 | mlflow The open source AI engineering platform for agents, LLMs, and ML models. View signal profile → | Growth | US | 211 | +21% | 100 | +0% | 2 | Framework migration |
| 2 | roboflowView signal profile → | Growth | US | 190 | +13% | 92 | +0% | 2 | Framework migration |
| 3 | modular Modular is an integrated, composable suite of tools that simplifies your AI infrastructure so your team can develop, dep View signal profile → | Growth | US | 514 | -12% | 100 | +0% | 0 | Framework migration |
| 4 | netdataView signal profile → | Growth | US | 70 | -18% | 100 | +14% | 1 | Framework migration |
Sorted by commit velocity change (14-day window, descending). Data last updated Q3 2025. Geography from GitHub org profiles.
The free Acceleration Watch: five venture-backed teams accelerating on the engineering signal, translated into plain English — 21 to 47 days before the deck circulates. No code-reading, no card.
In Q3 2025, mlflow leads ai & machine learning startups in United States with 211 commits over a 14-day window (+21% change) and 100 active contributors. Across all 4 tracked United States-based startups in this sector, the average 14-day commit velocity is 246 commits. The dominant signal pattern is "Framework migration", which typically indicates significant technical migration, which often precedes a pivot or platform upgrade.
United States accounts for 4 of the ai & machine learning startups in our tracking dataset for Q3 2025. This geographic view filters the broader sector rankings to help investors focused on United States-based deal flow identify engineering acceleration patterns within their target geography. Regional concentrations often reflect local regulatory environments, talent pools, and investor networks that shape startup trajectories differently from global averages.
We derive startup geography primarily from the GitHub organization profile location field, supplemented by a manually curated enrichment database of known startup headquarters. This means startups without a public GitHub location may appear in our global sector rankings but not in geographic filters. The geographic classification uses broad regions (United States, etc.) rather than city-level granularity to provide meaningful sample sizes for comparison.