Pareto

AI data research partner founded by Thiel Fellow Phoebe Yao, providing expert-curated training data for frontier AI models.

AI Training AI Data & Research
Visit Pareto
★ ★ ★ ★ ★ New (0 reviews)
Wilmington, USA HQ
573 employees

What is Pareto?

Pareto AI is a data research partner for leading AI companies, founded in 2020 by Phoebe Yao (Thiel Fellow and Forbes 30 Under 30 honoree), headquartered in Wilmington, Delaware with a San Francisco office. The company builds infrastructure for training next-generation AI models with a human-centric approach, providing expert-curated data for frontier models. What began as a bootcamp empowering women and work-at-home mothers for remote jobs evolved into a global platform focused on how expert knowledge drives AI's future. Services include advanced data labeling, recruitment of subject-matter experts, rigorous quality assurance, machine learning automation, deep domain-specific assessments, global data collection, iterative experiment design, and end-to-end rater management. Pareto emphasizes fair and ethical data labeling practices, ensuring workers are well-compensated with paid training. The company partners with research teams to design studies and deliver human-refined datasets.

Mission & values

Be the verification layer for frontier AI — turn nondeterministic expert judgment into durable reward signals and measure where models actually learn.

Qualifications

Pareto AI focuses on recruiting the "top 0.01% of expert labelers" to ensure the highest quality training data. For annotation roles, strong attention to detail, analytical thinking, and the ability to follow complex guidelines are essential. Domain expertise in specialized fields significantly enhances candidacy. The company offers paid training to familiarize workers with tasks and tools. Applicants undergo assessment tests to determine skill suitability for specific projects. For leadership and management roles, proven experience in people management and strategic thinking is required. No universal degree requirement exists, but expert-level knowledge in relevant domains is highly valued. The platform offers flexible, part-time, fully remote positions. Content creation, data labeling, and writing tasks form the core of available work.

Leadership

P

Phoebe Yao

Founder & CEO

Thiel Fellow and Forbes 30 Under 30 honoree. Pivoted Pareto into AI data labeling in 2023 after originally building a remote-work bootcamp.

Hiring process

  1. 1

    Apply at pareto.ai/careers

    Submit your application listing your domain expertise and credentials.

    About 1 day

  2. 2

    Vetting and qualification

    Pareto runs domain-skill assessments to filter for elite labelers / experts.

    About 5 days

  3. 3

    Onboarding and team match

    Get matched to a small specialist team aligned with an AI-research client need.

    About 3 days

  4. 4

    Project work

    RLHF preference data, expert evaluation, reward-signal design, and verification labeling.

Funding

StageSeed
Total raised$5,100,000
Investors
MaC Venture Capital Soma Capital Jonathan Swanson (Thumbtack) Seabed VC Fearless Fund Liquid2 Ventures Slope Agency

Awards & recognition

  • Founder is a Thiel Fellow and Forbes 30 Under 30 honoree

    Thiel Foundation / Forbes

Company information

Frequently asked questions

What does Pareto do today?
Pareto is the verification layer for reinforcement learning on real-world expertise — converting expert judgment into durable reward signals and measuring AI capability frontiers. Pivoted into AI data labeling in 2023.
Where is the contributor base concentrated?
Across 6 continents, with more than 30% of Pareto's active AI trainers based in Kenya.
How is the company funded?
$5.1M raised across seed rounds led by MaC Venture Capital and Soma Capital, with participation from Jonathan Swanson (Thumbtack), Seabed VC, Fearless Fund, Liquid2 Ventures, and Slope Agency.

Stay in the loop.

One email per week, 5 hand-picked roles.