Startup profile for Humyn Labs: latest funding, latest known valuation, total disclosed funding, investors, status, sources, and why the company is interesting. Part of the Venture Capital Tracker startup directory.

Startup profile · funding coverage

Humyn Labs funding, valuation and investors

Records and enriches real-world human task data and plans deployment test facilities for robotics developers.

Verified company research · funding, investors and latest developments

Keep track of Humyn Labs

Save this profile to your VCT watchlist for a quick return.

View watchlist

Funding, valuation & investors

Answer-first snapshot

Latest funding

Not publicly disclosed

Latest known valuation

Not publicly disclosed

Total disclosed equity funding

Not publicly disclosed

Excludes debt, grants, acquisitions, secondaries, and IPO proceeds.

Current status

private

series a · Bengaluru, Karnataka, India

Source links not recorded Last verified: 2026-11-18

Overview

Bengaluru-founded Humyn Labs supplies AI research teams and robotics vendors with structured, annotated data on humans performing tasks in real workplaces. Founders Manish Agarwal and Ishank Gupta launched the company in 2026. Its wearable/multimodal capture rigs record video, movement and other signals across 20+ countries; annotation and enrichment turn raw observations into data for embodied AI models.

Proposed deployment-readiness 'robot gyms' would test models with actual equipment and operators. The company says revenue recently surpassed a $50M annualized run rate; reporting explains this annualizes a small number of recent months, not audited trailing-year sales. The $47M Series A announcement includes $26M equity and $21M venture debt; only the equity portion represents primary equity funding. Valuation and customer contract rates undisclosed.

Why Humyn Labs is interesting

Its $47M November financing is two different kinds of capital: $26M equity led by Accel plus $21M venture debt led by Lighthouse Canton. Buyers need training data and a credible path from demos to repeatable robot operations.

Product & use cases

Customer is a robotics/embodied-AI lab needing representative training examples. Illustrative buyer workflow: (1) specify work tasks and sensing modalities; (2) commission field collection from human workers using Humyn capture hardware; (3) label hand-object interactions and movement traces with provenance; (4) deliver train/validation splits; (5) evaluate robot policies and failures; (6) use planned robot deployment centers to identify sim-to-real and safety gaps. Data rights, consent, geographical bias, annotation quality and safe worker/robot interaction require buyer diligence. Sales pricing and validation guarantees undisclosed.

  • Robotic grasping model: training examples of repeated pick-place actions recorded with motion/video (illustrative).
  • Warehouse robot developer: evaluate model failure modes against real aisle operations (illustrative; robot gym rollout planned).

Key facts

  • $47M disclosed package = $26M equity (Accel-led) + $21M venture debt (Lighthouse Canton-led); do not count as $47M equity.
  • Annualized revenue above $50M is a recent-month extrapolation, not a full-year booked revenue amount.
  • Reports 12 paying customers and physical-data collection in over 20 countries; independently audited dataset provenance not established.

Funding history (newest first)

Series A mixed financing: equity and venture debt

2026-11-18 $47M combined ($26M equity + $21M debt)
  • Accel (lead)
  • Prosus Ventures (participant)
  • BeeGlobal (participant)
  • Bajaj Finance/Finserv Intelligence (participant)
  • Dallas Venture Capital (participant)
  • Lighthouse Canton (participant)
  • Trifecta Capital (participant)

Source: Read source · business-standard.com

Investors in our directory

Funds linked from Humyn Labs's profile — open a fund page for stage focus and related deal articles.

Competitive landscape

  • Scale AI adjacent

    General model training-data services with some embodied AI applications; direct modality and rights comparisons needed.

  • Encord adjacent

    Multimodal labeling and dataset management, not necessarily equivalent human capture network.

  • NVIDIA Isaac Sim alternative

    Synthetic robot simulation complementary to, not interchangeable with, human field data.

  • RoboFlow adjacent

    Computer-vision dataset and tooling stack with different sensing emphasis.

Market / IPO context

Editorial / static context — not a live quote.

By Venture Capital Tracker

Last updated:

Editorial note: AI tools assisted with research, structure, or drafting. Venture Capital Tracker retains human editorial responsibility for factual accuracy, relevance, and source quality before publication.