Sama
Impact-sourcing data annotation platform blending automation with trained human annotators.
About
Sama (formerly Samasource) is a San Francisco training-data company founded in 2008 by Leila Janah on an impact-sourcing model: large annotation projects broken into small tasks performed by trained workers, primarily in Africa. It operates as a certified B Corporation and runs a secured cloud annotation platform covering data prep, task routing, ML-assisted labeling, QA and delivery, with a provider-claimed 99 percent client acceptance rate.
The current platform generation is built around three parts: SamaIQ for data insights and edge-case detection, SamaAssure for quality assurance, and SamaHub for project collaboration and reporting. Sama focuses on computer vision and sensor data, image, video and LiDAR annotation for automotive, AR/VR, agriculture and manufacturing clients. Caveat: the company has faced public criticism over working conditions and privacy practices in past press coverage, so diligence-minded buyers should review its current labor and data-handling policies.
What it can do
- Image, video and sensor data annotation and validation
- ML-assisted labeling with human-in-the-loop QA
- SamaIQ edge-case and data-distribution insights
- SamaHub project management and reporting
- Enterprise integrations and robust APIs
Setup
Talk to a Sama expert to scope a project; annotation runs on Sama's managed platform with a dedicated team.
Prerequisites
Scoped project brief; data-sharing agreement for regulated data.
Use cases
Works with
Trust score — 8/10
Scored 8/10 (verified tier, metrics-v1). Evidence: Strengths: founded 2008, certified B Corp, Platform 2.0 (SamaIQ/SamaAssure/SamaHub), public docs site, Wikipedia-backed company history; serves automotive, AR/VR, agriculture, manufacturing with image/video/sensor annotation; Docked for: (1) proprietary closed platform; (2) 99% acceptance and volume figures are provider-reported; (3) past press criticism of labor/privacy practices. Manual review: caveat on past working-conditions criticism noted in entry. License: proprietary (managed platform + workforce; project-based pricing).
How the trust score is built
Full methodology →- Adoption · stars/downloads, log-scaled vs peers
- 8.0/10
- Maintenance · recency, releases, issue load
- 8.0/10
- Openness · license + docs
- 5.0/10
- Manual review · sanity read + red-flag check
- 7.0/10
- · Strengths: founded 2008, certified B Corp, Platform 2.0 (SamaIQ/SamaAssure/SamaHub), public docs site, Wikipedia-backed company history
- · serves automotive, AR/VR, agriculture, manufacturing with image/video/sensor annotation
- · Docked for: (1) proprietary closed platform; (2) 99% acceptance and volume figures are provider-reported; (3) past press criticism of labor/privacy practices
- · manual review: caveat on past working-conditions criticism noted in entry
Sources
Every fact above was checked against these sources. Where a claim comes from the provider and couldn't be independently verified, the text says so.