Humyn Labs Raises $47M for Robotics Training Data

Original: India’s Humyn Labs raises $47M to feed U.S. robotics companies the data they need

Why This Matters

Physical-world training data is a hard bottleneck slowing humanoid and general robotics development.

Bengaluru-based Humyn Labs secured $47M in Series A funding—$26M equity led by Accel, $21M in venture debt—to supply physical-world training data to U.S. robotics and AI companies across 20+ countries.

Humyn Labs, founded in February 2026 by Manish Agarwal (former CEO of Nazara Technologies) and Ishank Gupta (former AB InBev executive), is building a data supply chain for robotics AI. The company developed proprietary two-camera hardware to record people performing everyday tasks, then processes and labels the footage so AI models can learn from it. Data is collected across 20+ countries, with Brazil, India, and South Korea as primary hubs—partly through the KGeN contributor network the founders built previously.

The $47M round splits into $26M equity (Accel leading, with Prosus Ventures, Bajaj Finance, Dallas Venture Capital, Jump Trading, and others) and $21M in venture debt via Lighthouse Canton and Trifecta Capital. The debt piece exists for a practical reason: Humyn pays data collectors daily or weekly but waits 30–60 days for AI lab customers to pay. Borrowing bridges that cash-flow gap without additional equity dilution.

Humyn reports 12 paying customers—frontier AI labs, robotics firms, and world model developers—with roughly 90% of revenue from the U.S. The company generated $10M in revenue from April to October 2026 and holds $20M in signed contracts, with another $20M in deals still pending. It competes with startups like Mecka AI, XDOF, and MicroAGI in the increasingly competitive physical AI data market.

Source

techcrunch.com — Read original →