Human expertise for the AI work that needs it most.
Recruit the people your model work actually requires, from broad consumer audiences to hard-to-find professionals and subject-matter experts. Use our platform, managed services, or custom recruiting to source, screen, and engage the right contributors.
The harder the task, the harder the right people are to find.
General-purpose data is easy to source. High-stakes AI work is different. It can require specific credentials, domain experience, geography, language, or software fluency, plus recruiting and screening that hold up under scrutiny.
Specialists are hard to reach
Clinicians, engineers, finance professionals, executives, and other experts are expensive and time-consuming to source manually, especially when the criteria get narrow.
Identity is hard to trust
Self-reported profiles are not enough for high-stakes work. Professional verification, screening, and fraud controls help reduce uncertainty before contributors ever reach your study.
Manual recruiting slows everything down
Complex recruiting can turn into spreadsheets, agencies, scheduling, and constant follow-up. A connected network and managed recruiting support keep the work moving.
Human data and expertise for AI training, testing, and evaluation.
Start with the audience you need and the work you need them to do. Recruit through our platform, add managed support when the project is complex, or use custom sourcing for specialists and low-incidence audiences.
Preference and ranking tasks
Recruit qualified contributors for side-by-side comparisons, ranking exercises, rubric-based reviews, and other structured feedback used in model and product evaluation.
Expert review and annotation
Bring in professionals and subject-matter experts to review outputs, complete structured tasks, or provide domain-specific feedback using criteria you define.
AI product and agent testing
Put AI experiences, copilots, and agents in front of the people who will use them to uncover usability issues, workflow gaps, and trust concerns.
Custom data collection
Run studies that collect voice, image, video, documents, device interactions, or other human-generated data, including in-person and other specialized projects.
Specialist and professional recruiting
Target by role, industry, seniority, skills, company size, product usage, and more, with managed sourcing available when your audience falls outside standard panel reach.
Repeat and longitudinal studies
Recontact participants or build recurring programs when you need to compare feedback over time, follow the same cohort, or support iterative AI development.
Start with people you can actually reach.
Our combined network includes broad consumer audiences, millions of professionals, and custom recruiting for harder-to-find specialists. Screening, professional verification, fraud controls, and managed recruiting help you build the right contributor pool for the work.
7.6M
People in the network
34K+
AI training sessions, 12 months
3.2M
Professionals across 140 industries
Your engagement stays scoped to your work from start to finish.
AI projects are commissioned and consented for the specific engagement. Customer data is kept isolated to that work rather than pooled, resold, or repurposed. For sensitive or specialized projects, our team can help define the recruiting and participation approach.
Engagement-level isolation
Customer prompts, materials, and study data stay within the engagement and are not pooled into a shared commercial dataset.
No model conflict
We do not build foundation models, so our role is to supply human input rather than compete with the systems we help evaluate.
Project-specific consent
Participants are recruited and consented for the work they are asked to complete, with additional handling available for sensitive studies.
Start with one audience and one use case.
Tell us who you need, what you need them to do, and how you plan to use the output. We can help determine the right mix of platform recruiting, managed services, and custom sourcing.
Frequently asked questions
Can’t find the answer you’re looking for? Talk to our team.
Aren't you a UX research company?
UserTesting and User Interviews have spent years building the infrastructure to recruit, screen, and connect companies with real people. AI training and evaluation extend that same capability into new types of work.
How do you qualify a specialist?
Qualification depends on the project. We can combine profile data, professional verification, custom screeners, work-email or LinkedIn signals, premium screening, and hands-on vetting for specialized recruits.
How do you support human evaluation?
We can support projects that collect structured human judgments and comparisons. The exact reporting, agreement method, adjudication process, and output format should be defined during scoping based on your evaluation design.
We run our own evals. Why add raters?
Human contributors can complement automated evals by reviewing the same outputs, surfacing context your grader may miss, and helping you understand where automated scores do or do not match human judgment.
What comes back at the end?
Deliverables depend on the project and method. They can include participant responses, recordings, structured study data, research outputs, or custom files agreed during scoping. API availability depends on the workflow.
What happens if quality falls short?
For managed and custom projects, acceptance criteria, replacement policies, and quality expectations can be defined up front. The exact remediation process depends on the scope and service model.
