Our mission
The people who train AI deserve to know what it pays.
A quiet workforce rates model answers, writes expert data, labels images, and red-teams frontier systems. They power every model you have ever used. Yet the map of who is hiring, and what each platform actually pays, has never existed in one place. AITraining.jobs is that map.
The market for AI training work exploded in a couple of years. 65 platforms now compete for talent (29 hiring right now): task-crowdsourcing sites, expert networks that pay PhDs and senior engineers by the hour, and enterprise annotation vendors. Each posts its own roles, sets its own rates, and pays on its own terms. For a worker, finding them means word of mouth, scattered subreddits, and a lot of guessing.
We built AITraining.jobs to replace the guessing with facts. We track the platforms, pull their open roles as they post, and compile pay benchmarks by role and platform from public sources. Then we put it behind a single search, so a trainer can see, in one place, where the work is and what it is worth, and get matched to the roles that fit their skills and their rate.
It comes down to one idea. Expert feedback, evaluation, and data labeling are the ground truth that every model is built on. The people doing that work should share in what they create, starting with the simple dignity of knowing what it is worth.
Principles
What we hold to
The pay is public, or it isn't real
Every rate on AITraining.jobs traces to a public source: a platform's own page, a job posting, or worker reports. Where a number is an estimate, we say so. We would rather show a range and its source than a confident figure we made up.
The worker is the customer
Most of this industry treats trainers as interchangeable supply. AITraining.jobs is built for the person doing the work: what pays best, who pays on time, and which platform actually fits your skills.
Fresh beats complete
Open roles change daily. We pull live from platform job boards and stamp every listing with when we last saw it, so you are never applying to a role that closed last month.
Who's behind it
A real person stands behind the numbers
AITraining.jobs publishes pay figures and platform picks, so it should be clear who compiles them and how to reach us.
Written by Yuma Heymans, Founder & CEO of AITraining.jobs.
Founder and CEO of AITraining.jobs and co-founder of the AI recruitment platform HeroHunt.ai, based in San Francisco, with years spent mapping how AI-training and data work is sourced, staffed, and paid.
Questions, a pay figure that looks off, or a platform we should add? Email Support and we will fix or add it. How we compile every pay figure is documented in our methodology.
Find your next AI training role.
Compare every platform, see what you could earn, and get matched to the best-paying work for your skills. Free for trainers.