What it sells
Help ML teams label the right data
Business model breakdown
Buyer: who owns the budget?
ML engineers and computer-vision teams / heads of ML; case studies include Eurocontrol, Harvard Medical School, Vale, Aigen, Kiwibot.
Trigger: what makes them buy now?
UnverifiedScaling computer-vision datasets where labeling everything is too costly; need to pick the right samples and speed model iteration.
Workflow: what inputs become what output?
Index data with automatic embeddings, filter and sample (diversity, dedup, outliers), label and QA with SAM3 auto-label plugins, evaluate against ground truth, export; LightlyTrain pretrains models without labels.
Economics: what it saves and how it charges
Software platform with open-source core (LightlySSL, LightlyStudio) plus commercial plans and AI data services; customers claim 2x faster model iteration and 50% less retraining time.
Distribution: how it reaches customers
Self-serve "Start for Free" signup, demo booking, open source and docs, case studies.
Funding history and coverage
| Round date | Stage | Amount | Investors | Source |
|---|---|---|---|---|
| — | Y Combinator Summer 2021 | — | Y Combinator | ycombinator.com |
Profiles summarize public reporting and company websites. Business-model notes are MaxBC's reading of that material, not confirmed facts; unknowns are left blank rather than guessed. Updated 2026-10-12.