Overall AI readiness score
A combined result across all audit signals.
Overall AI readiness score
A combined result across all audit signals.
Overall AI readiness assessment
AITraining.jobs is a legitimate job board for AI training work. Its llms.txt is accessible, valid, and consistent with the website. Approved for inclusion.
llms.txt is accessible and contains 13,800 tokens. No accessible llms-full.txt was detected, so deeper context still has to be assembled from regular pages. No separate ai.txt policy was detected; it is optional, but can clarify training, retrieval, and attribution preferences.
robots.txt is available. 11 of 11 tracked AI bots are not blocked. Declared sitemaps: 1. No Schema.org type was detected on the homepage. 0 OpenGraph tags were found and markup completeness is 0%, leaving more entity interpretation to crawlers.
The mobile Lighthouse profile adds Performance 88/100, Accessibility 97/100, Best Practices 100/100, SEO 100/100, and experimental Agentic Browsing 100/100. These signals have a limited weight: they complement rather than replace llms.txt, robots.txt, and structured-data checks.
A mobile Lighthouse measurement. Google’s experimental Agentic Browsing category is explained separately and does not replace the broader llmsmap AI-readiness score.
Performance
Accessibility
Best Practices
Technical SEO
Agentic Browsing
Mobile performance is 88/100, with the largest visible content block appearing in 2.7 s and the browser main thread blocked for 340 ms. Layout shift was 0. The main thread is where the browser runs JavaScript, calculates layout, and paints the page; long work there delays both user input and browser-agent actions.
Accessibility scored 97/100, Best Practices 100/100, and technical SEO 100/100. The experimental Agentic Browsing category scored 100/100. It measures signals Google currently tests for software agents and is shown separately from the llmsmap AI-readiness score.
Split long JavaScript tasks, defer non-critical scripts and styles, and shorten blocking request chains. This helps the primary content appear sooner and makes controls usable earlier.
Remove unused CSS and JavaScript, load heavy widgets on demand, and limit third-party scripts. Less code means less parsing and background work on the device.
Serve correctly sized modern formats, prioritise the primary visual, and lazy-load content below the first viewport.
Increase contrast for text, states, and interactive controls so both people and visual agents can distinguish content from actions and supporting labels.
First content
Main content
Layout stability
Blocking time
Visual speed
Machine-readable files, crawler policy, discovery, and homepage markup.
Full version was not found
ai.txt file was not found
1 sitemap found
Schema.org markup was not found on the homepage
OpenGraph tags were not found on the homepage
Based on robots.txt analysis
Declared discovery routes for crawlers and agents.
# AITraining.jobs
> Find AI training jobs and AI trainer jobs from every platform that pays people to train, evaluate, and label data for AI, with honest hourly pay measured from the live feeds and refreshed daily. Compare what the work pays by role, platform, and country.
AITraining.jobs is the live marketplace and pay-intelligence source for AI-training work: getting paid to train, evaluate, annotate, red-team, transcribe, and provide expert data for AI models. It tracks 63 platforms, pulls live openings daily from the 25 that publish a job feed, and compiles pay benchmarks for 12 kinds of work, with every figure labeled by how it is known (live from a job feed, platform-stated, third-party report, or estimate) and never presented as a guaranteed wage.
Live market as of 2026-10-14: 3,463 open AI-training roles across 25 platforms with live job feeds, 343 newly listed in the last 7 days (on platforms already tracked; a newly wired feed's back-catalog is not counted).
Hiring momentum: about 275 new AI-training roles are listed per week, averaged over the last 4 complete weeks (roles newly appearing on an already-tracked platform, so first wiring a feed's back-catalog is never counted as new hiring).
Highest-paying platforms right now by live measured median hourly rate: 1. Babel Audio $105/hr; 2. Mindrift $100/hr; 3. RemoExperts $95/hr; 4. Mercor $80/hr; 5. Sovrano AI $80/hr. Full ranking with methodology: https://aitraining.jobs/highest-paying-platforms.
## What this site provides
- A live feed of open AI-training roles pulled daily from 25 platforms' own job feeds.
- Honest pay benchmarks by role and by platform, spanning roughly $2 to $500 per hour.
- A pay-and-jobs hub for each kind of AI-training work (/roles/<category>): what it pays by platform, seniority, and country, who is hiring, and its live open roles.
- A plain-language education guide for each kind of AI-training work (/learn/<category>): what the work actually is day to day, who it suits, what it requires, and the steps to start.
- Country-aware filtering so a worker sees the roles they are actually eligible for.
## Pay by kind of AI-training work (live medians from today's open roles, with curated benchmarks for thin categories; live data as of 2026-10-14)
- AI Evaluation & Rating (RLHF): $25-$150/hr typical (live median $60/hr, measured from 172 distinct advertised hourly rates across 15 platforms; 91 of the 462 roles open here state no hourly rate). Guide: https://aitraining.jobs/roles/rlhf-eval
- Coding & Software Evaluation: $60-$145/hr typical (live median $65/hr, measured from 168 distinct advertised hourly rates across 12 platforms; 132 of the 364 roles open here state no hourly rate). Guide: https://aitraining.jobs/roles/coding
- Math & STEM: $40-$125/hr typical (live median $60/hr, measured from 262 distinct advertised hourly rates across 10 platforms; 83 of the 351 roles open here state no hourly rate). Guide: https://aitraining.jobs/roles/stem-math
- Law, Medicine & Finance: $65-$200/hr typical (live median $80/hr, measured from 475 distinct advertised hourly rates across 15 platforms; 183 of the 751 roles open here state no hourly rate). Guide: https://aitraining.jobs/roles/domain-expert
- Agentic Tasks & RL Environments: $30-$80/hr typical (live median $50/hr, measured from 16 distinct advertised hourly rates across 3 platforms; 20 of the 36 roles open here state no hourly rate). Project commissions: $20,000 to $300,000 per commissioned RL environment, project-based and distinct from the hourly rate. Guide: https://aitraining.jobs/roles/agentic-eval
- Data Annotation & Labeling: $13-$60/hr typical (live median $30/hr, measured from 62 distinct advertised hourly rates across 10 platforms; 46 of the 111 roles open here state no hourly rate). Guide: https://aitraining.jobs/roles/data-labeling
- Robotics & Physical-AI Data: $9-$105/hr typical (live median $14/hr, measured from 27 distinct advertised hourly rates across 5 platforms; 2 of the 30 roles open here state no hourly rate). Guide: https://aitraining.jobs/roles/robotics-data
- Multilingual & Localization: $20-$65/hr typical (live median $35/hr, measured from 477 distinct advertised hourly rates across 11 platforms; 306 of the 844 roles open here state no hourly rate). Guide: https://aitraining.jobs/roles/multilingual
- Transcription, Audio & Speech: $25-$70/hr typical (live median $30/hr, measured from 258 distinct advertised hourly rates across 11 platforms; 167 of the 429 roles open here state no hourly rate). Guide: https://aitraining.jobs/roles/audio-speech
- Writing & Content: $50-$130/hr typical (live median $60/hr, measured from 15 distinct advertised hourly rates across 3 platforms; 13 of the 31 roles open here state no hourly rate). Guide: https://aitraining.jobs/roles/writing
- Red-Teaming & AI Safety: $20-$100/hr typical (live median $50/hr, measured from 24 distinct advertised hourly rates across 5 platforms; 5 of the 33 roles open here state no hourly rate). Guide: https://aitraining.jobs/roles/red-teaming
- Research Studies & Surveys: $55-$135/hr from the 2 platforms that publish a rate for it (median $90/hr from 4 distinct advertised rates; 9 platforms hire for this work). Too few platforms publishing rates to call this a market rate; the compiled benchmark is $8–$40/hr for consumer surveys, $75–$150/hr equivalent for professional and expert panels. Guide: https://aitraining.jobs/roles/research-studies
## Key pages
- [Live open roles](https://aitraining.jobs/) - the daily-refreshed feed of open roles across the live platform job feeds.
- [All AI-training platforms](https://aitraining.jobs/platforms) - the full directory of every tracked platform, with its pay, work types, legitimacy, payout terms, and live open roles.
- [Every kind of AI-training work](https://aitraining.jobs/roles) - all role categories with live pay and who hires for each.
- [Pay explorer](https://aitraining.jobs/pay) - compare hourly pay across platforms and estimate your rate.
- [Highest-paying platforms](https://aitraining.jobs/highest-paying-platforms) - every platform ranked by the live measured median hourly rate its open roles advertise, refreshed daily, with the strongest payers first.
- [Machine-readable pay data](https://aitraining.jobs/pay.json) - today's live pay-by-work-type dataset as JSON (the DataDownload behind the /pay Dataset), free to reuse under CC BY 4.0; the same rows as a spreadsheet at https://aitraining.jobs/pay.csv.
- [The State of AI Training Pay report](https://aitraining.jobs/ai-training-pay-report) - the citable findings from that data (typical and median pay, the best-paid kinds of work and platforms, where the hiring is), with a ready citation; cite the date, the figures are re-measured daily.
- [Research: AI training labor market](https://aitraining.jobs/research) - live, citable reports built from the platforms' job boards and from apply clicks on this site, each with numbered figures, definitions and frozen monthly editions (cite an edition for figures that never change). The figures are on the pages, free to quote with credit; the data files are sent by email on request from each report's page.
- [Applicant Demand for AI Training Work](https://aitraining.jobs/research/applicant-demand) - From 16 Sep 2026 to 13 Oct 2026, people on AITraining.jobs made 5,307 apply clicks; per open role, interest was highest in Writing & Content (3.39x the average role) and lowest in Law, Medicine & Finance (0.54x).
- [AI Training Hiring Activity](https://aitraining.jobs/research/hiring-activity) - On 13 Oct 2026, 3,326 AI training roles were open across the job boards of 25 platforms. On the platforms tracked since 16 Sep 2026, open roles went down 6.5%, from 2,634 to 2,462; SuperAnnotate (SME), Turing, Anyone AI, Perle Labs, Sovrano AI, Pila8, RemoExperts and Babel Audio, first tracked during the window, added 864. 1,164 roles were newly listed and 1,351 removed in between.
- [AI Training Platforms Compared](https://aitraining.jobs/research/platform-scorecard) - On 13 Oct 2026, 3,326 AI training roles were open across the job boards of 25 platforms; OpenTrain AI listed the most (19.8%) and Babel Audio advertised the highest median pay ($105/hr).
- [Languages in Demand for AI Training](https://aitraining.jobs/research/languages-in-demand) - On 13 Oct 2026, open AI training roles named 95 languages; English, Spanish and French had the most openings, and Hebrew roles advertised the highest median pay ($55/hr).
- [Expert Fields in Demand for AI Training](https://aitraining.jobs/research/expert-fields-in-demand) - On 13 Oct 2026, 1,447 open AI training roles named an expert field; Software & computer science, Finance, accounting & economics and Medicine & nursing had the most openings, and Psychology & mental health advertised the highest median pay ($150/hr).
- [How AI Training Roles Advertise Pay](https://aitraining.jobs/research/pay-transparency) - On 13 Oct 2026, 68.6% of 3,326 open AI training roles stated an hourly rate, with a median of $55/hr and 53.7% at $50/hr or more.
- [Where AI Training Work Is Open](https://aitraining.jobs/research/where-ai-training-work-is-open) - On 13 Oct 2026, 28.6% of open AI training roles accepted applicants from any country; residents of the United States could apply to the most (2,077 of 3,326).
- [The AI Training Jobs Index](https://aitraining.jobs/research/ai-training-jobs-index) - From 3 Aug 2026 to 11 Oct 2026, open AI training roles on the platforms tracked throughout went up 10.4% (index 110.4); platforms listed 234 new roles in the last week and the advertised median stood at $58/hr.
- [What AI Training Roles Require](https://aitraining.jobs/research/what-ai-training-roles-require) - Of 1,434 open AI training roles with a full description on 13 Oct 2026, 22.4% mention a PhD or master's degree, 19.2% a professional credential and 22.1% an assessment or test.
- [How AI Training Applications Work](https://aitraining.jobs/research/how-ai-training-applications-work) - Across 1,149 application forms of AI training roles open on 13 Oct 2026, 93.4% require a resume, 80.7% ask for a LinkedIn profile and a typical form has 10 fields.
- [AI Training Pay Trends](https://aitraining.jobs/research/ai-training-pay-trends) - From the week of 31 Aug 2026 to the week of 5 Oct 2026, on the same platforms, 8 kinds of AI training work had a market rate at both ends; Data Annotation & Labeling moved most (+60%), and Law, Medicine & Finance paid the highest median ($80/hr).
- [AI Training Job Titles](https://aitraining.jobs/research/ai-training-job-titles) - On 13 Oct 2026, AI training roles were most often titled "Expert or SME", "Trainer" and "Specialist"; roles titled "Specialist" advertised the highest median pay ($65/hr).
- [What Raises AI Training Pay](https://aitraining.jobs/research/ai-training-pay-premiums) - Compared within each kind of work on 13 Oct 2026, a PhD in the description went with +22.0%, a professional credential in the description went with +19.6%, a profession named in the title went with +18.6% and a language named in the title with -25.9% in advertised pay.
- [AI Training Market Concentration](https://aitraining.jobs/research/ai-training-market-concentration) - On 13 Oct 2026, AI training listings were unconcentrated overall (HHI 1,024 across 25 platforms); Agentic Tasks & RL Environments was the most concentrated kind of work, led by OpenTrain AI with 80% of its open roles.
- [Data Quality and Coverage](https://aitraining.jobs/research/data-quality-and-coverage) - 5 of 7 independent cross-checks match exactly; the 25 job boards behind these reports were last read by 14 Oct 2026, 06:01 UTC, and 69% of open roles state an hourly rate.
- [How Long AI Training Roles Stay Open](https://aitraining.jobs/research/how-long-ai-training-roles-stay-open) - Half of the AI training roles listed from 3 Aug 2026 to 13 Oct 2026 came down within 25 days of being listed; Agentic Tasks & RL Environments turned over fastest (median 8.3 days), and counting only removed roles would put the median at 14 days.
- [When AI Training Roles Are Posted](https://aitraining.jobs/research/when-ai-training-roles-are-posted) - From 16 Sep 2026 to 13 Oct 2026, 84.8% of dated AI training postings fell on a weekday (Wednesday the busiest day), while people applied most on Monday and in the 15:00 UTC hour.
- [What Draws Applicants to AI Training Roles](https://aitraining.jobs/research/what-draws-applicants-to-ai-training-roles) - Compared within the same kind of work from 16 Sep 2026 to 13 Oct 2026, a listing under 14 days old went with 5.6 times the clicks per open role, 'freelance' in the title with 73% fewer and being open worldwide with 2.2 times: associations among the people who use AITraining.jobs, partly the site's own ranking of roles.
- [AI Training Work Without Experience](https://aitraining.jobs/research/ai-training-work-without-experience) - On 13 Oct 2026, 22.5% of open AI training roles (747) asked for no expert field, level or credential in their title and advertised a median $28/hr, and 36.8% of the roles with a full description mentioned no degree, years of experience or credential, though 16.1% of those still mentioned a test.
- [Programming Languages in AI Training](https://aitraining.jobs/research/programming-languages-in-ai-training) - Of 352 Coding & Software Evaluation roles open on 13 Oct 2026, 54.0% name a programming language, Python most often (25.6%); where enough roles on two or more platforms state pay, median advertised rates run from $60/hr (Kotlin) to $100/hr (R).
- [What AI Trainers Do](https://aitraining.jobs/research/what-ai-trainers-do) - Of 1,434 open AI training roles with a full description on 13 Oct 2026 (43.1% of open roles, at the expert networks that publish descriptions), 52.2% mention writing prompts, ideal answers or evaluation tasks, 53.5% rating, scoring or ranking model answers, 19.6% recording or transcribing audio and video, 14.4% writing or reviewing code and 5.6% red-teaming.
- [On-Site AI Training Work](https://aitraining.jobs/research/on-site-ai-training-work) - On 13 Oct 2026, 2.7% of 3,326 open AI training roles had to be done on-site or hybrid, 86.4% were remote and 10.9% did not say; the largest group of on-site roles was data labelling and quality work, and most were in the United States, the United Kingdom and India.
- [How AI Training Work Is Paid](https://aitraining.jobs/research/how-ai-training-work-is-paid) - On 13 Oct 2026, 97.2% of the 2,352 open AI training roles quoting pay quoted it by the hour, 85.8% of roles stating an engagement type were contract or freelance, and among roles with a full description 9.3% paid per task or other output while 7.1% stated a payout schedule.
- [The AI Training Rate Card](https://aitraining.jobs/research/ai-training-rate-card) - On 13 Oct 2026, for roles open to US residents, the best-paying platform advertised 2.88 times the rate of the worst-paying one in the typical kind of AI training work, and up to 6.2 times in AI Evaluation & Rating (RLHF) ($93 at RemoExperts against $15 at Turing).
- [Which AI Labs Are Hiring for Human Data](https://aitraining.jobs/research/ai-labs-human-data-hiring) - On 14 Oct 2026, 16 of 25 AI labs with a public job board had 101 roles open on the teams that buy, run and evaluate human data, xAI the most (27).
- [Where Human-Data Vendors Are Building](https://aitraining.jobs/research/human-data-vendors-hiring) - On 14 Oct 2026, 14 human-data vendors had 660 staff roles open on their own boards, Scale AI the most (183); engineering and product is the largest function.
- [Who Is Buying AI Training Data](https://aitraining.jobs/research/who-is-buying-ai-training-data) - On 14 Oct 2026, 14 roles at AI labs and human-data vendors had the job of sourcing, licensing or acquiring training data, the most at OpenAI, and 20 more were building RL environments: the desks data owners and brokers sell to.
- [What AI Labs Pay for Human Data](https://aitraining.jobs/research/what-ai-labs-pay-for-human-data) - 59 published figures on what AI labs pay for human data, from 2021 to 11 Oct 2026: lab spend, vendor bill rates against worker pay, vendors' take, RL-environment prices, prices per task and contract sizes; the largest spend figure recorded: Surge reportedly generated $1.2 billion in revenue in 2024 from work with AI labs (Surge AI, reported).
- [What AI Labs Are Buying](https://aitraining.jobs/research/what-ai-labs-are-buying) - To 11 Oct 2026, the evidence shows labs buying more coding (production-like environments), computer-use app and website clones and egocentric and robotics demonstration data and less generalist annotation and microtasks and static maths and verifiable coding tasks.
- [AI Training Data Licensing Deals](https://aitraining.jobs/research/ai-training-data-licensing-deals) - From 11 Jul 2023 to 11 Oct 2026, 89 content and data licensing deals between AI companies and data owners are recorded, 22 with a published price; the largest is News Corp with OpenAI ("Could be" close to $250 million over five years, per The Wall Street Journal (as reported by VentureBeat); terms not disclosed by the parties).
- [Lawsuits and Settlements Over AI Training Data](https://aitraining.jobs/research/ai-training-data-lawsuits) - From 20 Sep 2023 to 11 Oct 2026, 68 lawsuits, rulings, settlements and regulatory actions over AI training data are recorded; the largest amount is Bartz v. Anthropic ($1.5 billion, approximately $3,000 per work).
- [Company Data Brokerage: Buyers, Prices and Terms](https://aitraining.jobs/research/company-data-brokerage) - Companies' records (email, chat, tickets, code, CRM) and the archives of defunct companies now sell to AI labs and principal vendors as raw material for RL environments: 62 prices, programs and terms recorded to 11 Oct 2026, up to $12,500,000 for a single archive and a $1 billion over the next 12 months buying program at micro1.
- [Human Data Quality, Fraud and Security](https://aitraining.jobs/research/human-data-quality-and-fraud) - From 13 Jun 2023 to 11 Oct 2026, 41 studies, incidents and vendor statements document the quality risks of human data: workers using AI models to do the tasks (33-46% of crowd workers (range estimate) in the first measurement), fake and resold accounts, quality failures on lab projects and breaches at vendors.
- [AI Training Pay Against the Local Minimum Wage](https://aitraining.jobs/research/ai-training-pay-vs-minimum-wage) - On 13 Oct 2026, the median advertised AI training rate open to residents of 19 countries ran from 3.3 times the local minimum wage (Australia) to 238.2 times (India).
- [Where AI Labs Are Pushing: The Expertise Being Hired, Week by Week](https://aitraining.jobs/research/where-labs-are-pushing) - From the week of 3 Aug 2026 to the week of 5 Oct 2026, on the 15 platforms tracked throughout, Medicine & nursing grew fastest among expert fields (+76.4% open roles), Spanish among languages (+59.4%) and Transcription, Audio & Speech among kinds of work (+63.1%).
- [AI Training Vendor Health](https://aitraining.jobs/research/vendor-health-panel) - From 16 Sep 2026 to 13 Oct 2026, open roles on the 17 AI training platforms tracked throughout went down 6.5% with re-key batches set aside: 6 grew and 5 shrank by 10% or more while 4 held steady, and the largest moves in roles were Mercor (-163 roles, -46.3%) and Alignerr (+27, +71.1%).
- [The Expert Scarcity Index for AI Training](https://aitraining.jobs/research/expert-scarcity-index) - On 13 Oct 2026, Chemistry was the hardest expert field to recruit for AI training work (scarcity index 1.74x, 1.00 = the market), German the hardest language (1.56x) and Math & STEM the hardest kind of work (1.38x); 7 groups were scarcer than the market on all three measures: applicants per opening, time to close and pay.
- [Real-World Data Collection Work](https://aitraining.jobs/research/real-world-data-collection-work) - On 13 Oct 2026, 67 roles on 9 AI training platforms recruited people to collect real-world data for robots and physical AI, TransPerfect DataForce the most (35), at a median advertised $16 an hour; 38.8% had to be done at a site, and 4 AI labs listed 13 robot data-collection roles on their own boards.
- [How AI Training Platforms Screen Experts](https://aitraining.jobs/research/how-ai-training-platforms-screen-experts) - Of the 1,434 open roles whose full description is read (11 platforms), 25.1% state a test, interview or check, most often an assessment, test or trial task (18.3%), and 19.2% name a professional credential or licence; the 1,149 application forms read on 11 platforms require a median of 10 fields, and 93.4% require a resume.
- [Funding and Valuations of AI Training Data Companies](https://aitraining.jobs/research/ai-training-industry-funding) - From 1 Jan 2024 to 10 Oct 2026, the companies that supply human data to AI labs disclosed $3.1B in 38 funding rounds, and Scale AI carries the highest valuation at $29B.
- [Revenue and Scale of AI Training Data Companies](https://aitraining.jobs/research/ai-training-industry-revenue) - The listed human-data companies report their revenue in filings and the private ones through statements and press reports; the largest private figure recorded is Mercor's over $2 billion in gross revenue run rate (run-rate Jun 2026), and Innodata's revenue rose 47.6% in FY2025.
- [Lawsuits, Regulation and Labour Actions in AI Training Work](https://aitraining.jobs/research/ai-training-work-legal-tracker) - From 1 Jan 2023 to 10 Oct 2026, 30 lawsuits, rulings, investigations, laws and labour actions concerned the people who train AI; Scale AI is named in the most.
- [What Studies Find About Pay in AI Data Work](https://aitraining.jobs/research/research-on-ai-data-work-pay) - Studies of data work have measured hourly earnings as low as an average of a little more than 90 cents an hour (How the AI industry profits from catastrophe) and a median of $3.31 across 26 measured figures, while the AI training roles open today advertise a median of $55 an hour, most of it expert work.
- [The Market for AI Evaluation & Rating (RLHF) Work](https://aitraining.jobs/research/ai-evaluation-rlhf-market) - On 13 Oct 2026, 448 roles in AI Evaluation & Rating (RLHF) were open on 18 platforms (SuperAnnotate (SME) listed 43.5%); their median advertised rate was $45/hr, 18% below the $55/hr of all AI training work; each open role drew 2.05x the apply clicks of the average role.
- [The Market for Coding & Software Evaluation Work](https://aitraining.jobs/research/coding-ai-training-market) - On 13 Oct 2026, 352 roles in Coding & Software Evaluation were open on 14 platforms (OpenTrain AI listed 24.7%); their median advertised rate was $65/hr, 18% above the $55/hr of all AI training work; each open role drew 0.87x the apply clicks of the average role.
- [The Market for Math & STEM Work](https://aitraining.jobs/research/math-stem-ai-training-market) - On 13 Oct 2026, 347 roles in Math & STEM were open on 14 platforms (Anyone AI listed 26.8%); their median advertised rate was $60/hr, 9% above the $55/hr of all AI training work; each open role drew 0.59x the apply clicks of the average role.
- [The Market for Law, Medicine & Finance Work](https://aitraining.jobs/research/law-medicine-finance-ai-training-market) - On 13 Oct 2026, 701 roles in Law, Medicine & Finance were open on 19 platforms (OpenTrain AI listed 30.5%); their median advertised rate was $75/hr, 36% above the $55/hr of all AI training work; each open role drew 0.54x the apply clicks of the average role.
- [The Market for Agentic Tasks & RL Environments Work](https://aitraining.jobs/research/agentic-rl-environment-market) - On 13 Oct 2026, 35 roles in Agentic Tasks & RL Environments were open on 3 platforms (OpenTrain AI listed 80.0%); their median advertised rate was $53/hr, 4% below the $55/hr of all AI training work; each open role drew 2.61x the apply clicks of the average role.
- [The Market for Data Annotation & Labeling Work](https://aitraining.jobs/research/data-annotation-market) - On 13 Oct 2026, 107 roles in Data Annotation & Labeling were open on 14 platforms (Welo Data listed 41.1%); their median advertised rate was $28/hr, 49% below the $55/hr of all AI training work; each open role drew 2.32x the apply clicks of the average role.
- [The Market for Robotics & Physical-AI Data Work](https://aitraining.jobs/research/robotics-physical-ai-data-market) - On 13 Oct 2026, 30 roles in Robotics & Physical-AI Data were open on 7 platforms (micro1 listed 40.0%); their median advertised rate was $15/hr, 73% below the $55/hr of all AI training work; each open role drew 2.19x the apply clicks of the average role.
- [The Market for Multilingual & Localization Work](https://aitraining.jobs/research/multilingual-ai-training-market) - On 13 Oct 2026, 810 roles in Multilingual & Localization were open on 16 platforms (Invisible listed 30.0%); their median advertised rate was $37/hr, 33% below the $55/hr of all AI training work; each open role drew 0.64x the apply clicks of the average role.
- [The Market for Transcription, Audio & Speech Work](https://aitraining.jobs/research/transcription-audio-speech-market) - On 13 Oct 2026, 416 roles in Transcription, Audio & Speech were open on 15 platforms (no platform listed more than 14.4%); their median advertised rate was $30/hr, 46% below the $55/hr of all AI training work; each open role drew 1.19x the apply clicks of the average role.
- [The Market for Writing & Content Work](https://aitraining.jobs/research/writing-ai-training-market) - On 13 Oct 2026, 31 roles in Writing & Content were open on 10 platforms (OpenTrain AI listed 32.3%); their median advertised rate was $60/hr, 9% above the $55/hr of all AI training work; each open role drew 3.39x the apply clicks of the average role.
- [The Market for Red-Teaming & AI Safety Work](https://aitraining.jobs/research/red-teaming-ai-safety-market) - On 13 Oct 2026, 29 roles in Red-Teaming & AI Safety were open on 7 platforms (Mercor listed 15 of them); their median advertised rate was $55/hr, about the same as the $55/hr of all AI training work; each open role drew 1.05x the apply clicks of the average role.
- [The Market for Research Studies & Surveys Work](https://aitraining.jobs/research/research-studies-surveys-market) - On 13 Oct 2026, 20 roles in Research Studies & Surveys were open on 3 platforms (TransPerfect DataForce listed 12 of them); 3 of them stated an hourly rate; each open role drew 1.62x the apply clicks of the average role.
- [OpenTrain AI Hiring Data](https://aitraining.jobs/research/opentrain-ai-hiring-data) - On 13 Oct 2026, OpenTrain AI listed 660 open AI training roles (19.8% of the market, the largest of 25 platforms), 32.4% of them in Law, Medicine & Finance; 27.7% stated an hourly rate, at a median of $48/hr, and 75.0% accepted applicants only from listed countries, most often the United States.
- [Invisible Hiring Data](https://aitraining.jobs/research/invisible-technologies-hiring-data) - On 13 Oct 2026, Invisible listed 531 open AI training roles (16.0% of the market, the 2nd largest of 25 platforms), 45.8% of them in Multilingual & Localization; 99.2% stated an hourly rate, at a median of $65/hr, and 63.8% accepted applicants only from listed countries, most often the United States.
- [micro1 Hiring Data](https://aitraining.jobs/research/micro1-hiring-data) - On 13 Oct 2026, micro1 listed 230 open AI training roles (6.9% of the market, the 5th largest of 25 platforms), 34.8% of them in Law, Medicine & Finance; 98.7% stated an hourly rate, at a median of $73/hr, and 68.3% accepted applicants from any country.
- [Welo Data Hiring Data](https://aitraining.jobs/research/welo-data-hiring-data) - On 13 Oct 2026, Welo Data listed 231 open AI training roles (6.9% of the market, the 4th largest of 25 platforms), 48.1% of them in Multilingual & Localization; 77.5% stated an hourly rate, at a median of $20/hr, and 98.3% accepted applicants only from listed countries, most often the United States.
- [AfterQuery Hiring Data](https://aitraining.jobs/research/afterquery-hiring-data) - On 13 Oct 2026, AfterQuery listed 208 open AI training roles (6.3% of the market, the 6th largest of 25 platforms), 55.3% of them in Law, Medicine & Finance; all stated an hourly rate, at a median of $75/hr, and 85.1% accepted applicants worldwide except listed countries.
- [Mercor Hiring Data](https://aitraining.jobs/research/mercor-hiring-data) - On 13 Oct 2026, Mercor listed 189 open AI training roles (5.7% of the market, the 8th largest of 25 platforms), 24.3% of them in Math & STEM; 79.4% stated an hourly rate, at a median of $80/hr, and 55.6% accepted applicants from any country.
- [Sigma AI Hiring Data](https://aitraining.jobs/research/sigma-ai-hiring-data) - On 13 Oct 2026, Sigma AI listed 88 open AI training roles (2.6% of the market, the 10th largest of 25 platforms), 92.0% of them in Multilingual & Localization; none stated an hourly rate, and all accepted applicants only from listed countries, most often the United Kingdom.
- [Braintrust Hiring Data](https://aitraining.jobs/research/braintrust-hiring-data) - On 13 Oct 2026, Braintrust listed 75 open AI training roles (2.3% of the market, the 11th largest of 25 platforms), 53.3% of them in Multilingual & Localization; 72.0% stated an hourly rate, at a median of $25/hr, and 98.7% accepted applicants only from listed countries, most often the United States.
- [Alignerr Hiring Data](https://aitraining.jobs/research/alignerr-hiring-data) - On 13 Oct 2026, Alignerr listed 65 open AI training roles (2.0% of the market, the joint 12th largest of 25 platforms), 70.8% of them in Transcription, Audio & Speech; all stated an hourly rate, at a median of $25/hr, and all accepted applicants from any country.
- [TransPerfect DataForce Hiring Data](https://aitraining.jobs/research/transperfect-dataforce-hiring-data) - On 13 Oct 2026, TransPerfect DataForce listed 58 open AI training roles (1.7% of the market, the 14th largest of 25 platforms), 36.2% of them in Multilingual & Localization; none stated an hourly rate, and all accepted applicants only from listed countries, most often the United States.
- [Appen Hiring Data](https://aitraining.jobs/research/appen-hiring-data) - On 13 Oct 2026, Appen listed 48 open AI training roles (1.4% of the market, the 15th largest of 25 platforms), 66.7% of them in Transcription, Audio & Speech; 29.2% stated an hourly rate, at a median of $35/hr, and 79.2% accepted applicants only from listed countries, most often the United States.
- [Mindrift Hiring Data](https://aitraining.jobs/research/mindrift-hiring-data) - On 13 Oct 2026, Mindrift listed 29 open AI training roles (0.9% of the market, the 18th largest of 25 platforms), all in Law, Medicine & Finance; all stated an hourly rate, at a median of $37/hr, and all accepted applicants only from listed countries, most often the United States.
- [xAI (SpaceXAI) Hiring Data](https://aitraining.jobs/research/xai-hiring-data) - On 13 Oct 2026, xAI (SpaceXAI) listed 17 open AI training roles (0.5% of the market, the 21st largest of 25 platforms), 76.5% of them in Multilingual & Localization; 94.1% stated an hourly rate, at a median of $40/hr, and 94.1% accepted applicants from any country.
- [Terac Hiring Data](https://aitraining.jobs/research/terac-hiring-data) - On 13 Oct 2026, Terac listed 20 open AI training roles (0.6% of the market, the 19th largest of 25 platforms), 30.0% of them in Research Studies & Surveys; 65.0% stated an hourly rate, at a median of $75/hr, and all accepted applicants only from listed countries, most often the United States.
- [Machine-readable open-roles market](https://aitraining.jobs/roles.json) - today's live open-role counts by kind of work and by platform, plus feed freshness, as JSON (the DataDownload behind the /roles Dataset), free to reuse under CC BY 4.0.
- [Data reference](https://aitraining.jobs/data) - every field of /roles.json, /pay.json and /pay.csv, the terms (CC BY 4.0, readable by code with no key, CORS enabled, refreshed daily; for people the files are sent by email) and how to cite them.
- [How we compile pay](https://aitraining.jobs/methodology) - exactly how every pay figure is sourced, labeled, and kept current.
- [How it works](https://aitraining.jobs/how-it-works) - the platform directory and the sourcing method.
- [AI Evaluation & Rating (RLHF) pay & jobs](https://aitraining.jobs/roles/rlhf-eval)
- [Coding & Software Evaluation pay & jobs](https://aitraining.jobs/roles/coding)
- [Math & STEM pay & jobs](https://aitraining.jobs/roles/stem-math)
- [Law, Medicine & Finance pay & jobs](https://aitraining.jobs/roles/domain-expert)
- [Agentic Tasks & RL Environments pay & jobs](https://aitraining.jobs/roles/agentic-eval)
- [Data Annotation & Labeling pay & jobs](https://aitraining.jobs/roles/data-labeling)
- [Robotics & Physical-AI Data pay & jobs](https://aitraining.jobs/roles/robotics-data)
- [Multilingual & Localization pay & jobs](https://aitraining.jobs/roles/multilingual)
- [Transcription, Audio & Speech pay & jobs](https://aitraining.jobs/roles/audio-speech)
- [Writing & Content pay & jobs](https://aitraining.jobs/roles/writing)
- [Red-Teaming & AI Safety pay & jobs](https://aitraining.jobs/roles/red-teaming)
- [Research Studies & Surveys pay & jobs](https://aitraining.jobs/roles/research-studies)
## Job lists for specific searches (today's open roles, updated daily)
- [Remote AI training jobs](https://aitraining.jobs/remote-ai-training-jobs) - today's open remote roles (most of this work is remote), by kind of work, each linking its full job posting where the site holds one.
- [Full job descriptions](https://aitraining.jobs/jobs) - every open role whose complete posting the site holds, with its pay and the countries it is open to.
- [Remote AI training jobs for lawyers](https://aitraining.jobs/remote-attorney-jobs) - open roles for attorneys and legal experts.
- [Remote AI training jobs for clinicians](https://aitraining.jobs/remote-healthcare-jobs) - open roles for physicians, nurses and other clinicians.
- [Remote AI training jobs in finance](https://aitraining.jobs/remote-finance-jobs) - open roles for accountants, analysts and finance experts.
- [Voice acting and voice over jobs](https://aitraining.jobs/voice-acting-jobs) - open recording and voice roles for AI training.
- [How to get paid to train AI](https://aitraining.jobs/get-paid-to-train-ai) - what this work pays, which platforms are legit and how to start.
## AI training jobs by country (live roles open to applicants in each country)
- [AI training jobs by country](https://aitraining.jobs/countries) - how many live roles accept applicants in each listed country, and how to see yours on the job board.
- [AI training jobs in Australia](https://aitraining.jobs/countries/australia)
- [AI training jobs in Bangladesh](https://aitraining.jobs/countries/bangladesh)
- [AI training jobs in Canada](https://aitraining.jobs/countries/canada)
- [AI training jobs in Egypt](https://aitraining.jobs/countries/egypt)
- [AI training jobs in Germany](https://aitraining.jobs/countries/germany)
- [AI training jobs in Ghana](https://aitraining.jobs/countries/ghana)
- [AI training jobs in India](https://aitraining.jobs/countries/india)
- [AI training jobs in Indonesia](https://aitraining.jobs/countries/indonesia)
- [AI training jobs in Japan](https://aitraining.jobs/countries/japan)
- [AI training jobs in Kenya](https://aitraining.jobs/countries/kenya)
- [AI training jobs in Malaysia](https://aitraining.jobs/countries/malaysia)
- [AI training jobs in Nigeria](https://aitraining.jobs/countries/nigeria)
- [AI training jobs in Pakistan](https://aitraining.jobs/countries/pakistan)
- [AI training jobs in the Philippines](https://aitraining.jobs/countries/philippines)
- [AI training jobs in South Africa](https://aitraining.jobs/countries/south-africa)
- [AI training jobs in Spain](https://aitraining.jobs/countries/spain)
- [AI training jobs in the UK](https://aitraining.jobs/countries/united-kingdom)
- [AI training jobs in the USA](https://aitraining.jobs/countries/united-states)
## In Spanish, Portuguese and Japanese (AI training jobs open in Latin America, Spain and Japan)
- [Trabajos de entrenamiento de IA](https://aitraining.jobs/es) - Spanish: today's open AI-training roles, the ones that ask for Spanish, what they pay in USD and how each platform pays.
- [Trabajos de entrenamiento de IA en México](https://aitraining.jobs/es/mexico) - Spanish: the roles open to applicants in México.
- [Trabajos de entrenamiento de IA en Colombia](https://aitraining.jobs/es/colombia) - Spanish: the roles open to applicants in Colombia.
- [Trabajos de entrenamiento de IA en Argentina](https://aitraining.jobs/es/argentina) - Spanish: the roles open to applicants in Argentina.
- [Trabajos de entrenamiento de IA en España](https://aitraining.jobs/es/espana) - Spanish: the roles open to applicants in España.
- [Trabajos de entrenamiento de IA en Perú](https://aitraining.jobs/es/peru) - Spanish: the roles open to applicants in Perú.
- [Trabajos de entrenamiento de IA en Chile](https://aitraining.jobs/es/chile) - Spanish: the roles open to applicants in Chile.
- [Trabajos de entrenamiento de IA en Venezuela](https://aitraining.jobs/es/venezuela) - Spanish: the roles open to applicants in Venezuela.
- [Trabajos de entrenamiento de IA en Ecuador](https://aitraining.jobs/es/ecuador) - Spanish: the roles open to applicants in Ecuador.
- [Vagas de treinamento de IA no Brasil](https://aitraining.jobs/pt) - Portuguese: the roles open to applicants in Brazil, what they pay and how to get paid.
- [日本から応募できるAIトレーニングの仕事](https://aitraining.jobs/ja) - Japanese: the roles open to applicants in Japan, the ones in Japanese, pay in dollars and yen, how to get paid and whether the income must be declared.
## Alternatives to AI-training platforms (sites like each, with today's open roles)
- [AI training platform alternatives](https://aitraining.jobs/alternatives) - every alternatives page on the site.
- [Data Annotation Tech alternatives](https://aitraining.jobs/alternatives/dataannotation) - sites like DataAnnotation that hire for the same work, with their pay, who can join and their open roles today.
- [Outlier AI alternatives](https://aitraining.jobs/alternatives/outlier-scale-ai) - sites like Outlier (Scale AI) that hire for the same work, with their pay, who can join and their open roles today.
- [Mercor alternatives](https://aitraining.jobs/alternatives/mercor) - sites like Mercor that hire for the same work, with their pay, who can join and their open roles today.
- [Remotasks alternatives](https://aitraining.jobs/alternatives/remotasks) - sites like Remotasks that hire for the same work, with their pay, who can join and their open roles today.
- [TELUS International AI alternatives](https://aitraining.jobs/alternatives/telus-international-ai-community) - sites like TELUS Digital AI that hire for the same work, with their pay, who can join and their open roles today.
- [Appen alternatives](https://aitraining.jobs/alternatives/appen) - sites like Appen that hire for the same work, with their pay, who can join and their open roles today.
- [Alignerr alternatives](https://aitraining.jobs/alternatives/alignerr) - sites like Alignerr that hire for the same work, with their pay, who can join and their open roles today.
- [Clickworker alternatives](https://aitraining.jobs/alternatives/clickworker) - sites like Clickworker that hire for the same work, with their pay, who can join and their open roles today.
- [MTurk alternatives](https://aitraining.jobs/alternatives/amazon-mechanical-turk) - sites like Amazon Mechanical Turk that hire for the same work, with their pay, who can join and their open roles today.
## AI-training platforms (is it legit, what it pays, and its live roles or how to join)
- [OpenTrain AI: is it legit, what it pays, and live roles](https://aitraining.jobs/platforms/opentrain-ai) - $15–$120/hr; 698 open roles right now. Live median $50/hr across this platform's 207 advertised rates. Top roles: Domain expert / SME AI Trainer $40–$110/hr; Coding / code reviewer $50–$100+/hr. Payout: Stripe payouts.
- [Invisible: is it legit, what it pays, and live roles](https://aitraining.jobs/platforms/invisible-technologies) - $6–$65/hr; 531 open roles right now. Live median $65/hr across this platform's 527 advertised rates. Top roles: Expert domain specialist (AI Trainer) $6–$65/hr; Language / dialect specialist $8–$65/hr. Payout: Independent contractor; typically paid twice monthly.
- [SuperAnnotate (SME): is it legit, what it pays, and live roles](https://aitraining.jobs/platforms/superannotate) - $17–$115+/hr by domain and geography; 380 open roles right now. Top roles: Legal expert / attorney $40–$115/hr; Medical / doctor expert $40–$90/hr. Payout: Via Deel (contracts, tax/compliance).
- [micro1: is it legit, what it pays, and live roles](https://aitraining.jobs/platforms/micro1) - $10–$95/hr; frontier AI/ML to $200/hr; 244 open roles right now. Live median $70/hr across this platform's 240 advertised rates. Top roles: AI / ML engineer $70–$200/hr; Legal / STEM expert $40–$150/hr. Payout: USD via Deel (bank, PayPal, Wise, Payoneer).
- [Welo Data: is it legit, what it pays, and live roles](https://aitraining.jobs/platforms/welo-data) - Varies by role; see advertised rates; 233 open roles right now. Live median $20/hr across this platform's 181 advertised rates. Top roles: Data annotation & labeling Per project (undisclosed); Model output ranking / preference Per project (undisclosed). Payout: Per project through the Welo Works platform; workers are onboarded as either employees or freelancers depending on project. Exact payout method and cadence are not publicly documented.
- [Mercor: is it legit, what it pays, and live roles](https://aitraining.jobs/platforms/mercor) - $25–$75/hr typical; experts to $250/hr; 210 open roles right now. Live median $80/hr across this platform's 171 advertised rates. Top roles: Medical / healthcare expert $110–$250/hr; RL environment / agentic data $75–$200+/hr. Payout: Weekly via Stripe (1099 contractors).
- [AfterQuery: is it legit, what it pays, and live roles](https://aitraining.jobs/platforms/afterquery) - $45–$100/hr; experts to $175/hr; 208 open roles right now. Live median $75/hr across this platform's 208 advertised rates. Top roles: Math & STEM expert $88–$115/hr; Credentialed specialist (finance/law/med) $70–$108/hr. Payout: Weekly (Fridays) via Stripe to bank.
- [Turing: is it legit, what it pays, and live roles](https://aitraining.jobs/platforms/turing) - $15–$150/hr; 196 open roles right now. Live median $15/hr across this platform's 15 advertised rates. Top roles: Code generation reviewer $50–$100+/hr; Domain expert (finance/legal/med/STEM) $50–$100+/hr. Payout: Monthly for the hours you log: paid on the 5th calendar day of each month, through Turing's own portal or Deel depending on the project.
- [Anyone AI: is it legit, what it pays, and live roles](https://aitraining.jobs/platforms/anyone-ai) - $15-$150/hr (most roles $40-$80); 128 open roles right now. Live median $40/hr across this platform's 128 advertised rates. Top roles: Software engineering AI trainer (full-stack, backend, Python) $45-$80/hr; STEM expert, AI trainer (biology, chemistry, physics, math) up to $40/hr. Payout: Not published. Software roles are paid per project (hourly rates are estimates); no payout method or schedule is stated.
- [Sigma AI: is it legit, what it pays, and live roles](https://aitraining.jobs/platforms/sigma-ai) - ~$19–$51/hr; 90 open roles right now. Top roles: Linguistic annotator / labeler ~$41–$51/hr (est.); Transcriber (freelance) ~$19–$31/hr. Payout: Invoice-based commercial contract.
- [Braintrust: is it legit, what it pays, and live roles](https://aitraining.jobs/platforms/braintrust) - $20–$200/hr; 77 open roles right now. Live median $25/hr across this platform's 56 advertised rates. Top roles: AI code review $75–$200/hr; Expert evaluation / domain reviewer $60–$150/hr. Payout: Weekly; talent keeps 100% (no talent-side fee).
- [Mindrift: is it legit, what it pays, and live roles](https://aitraining.jobs/platforms/mindrift) - $15–$100+/hr; 70 open roles right now. Live median $100/hr across this platform's 29 advertised rates. Top roles: Business / strategy consultant up to $100/hr; Coding / SWE (agent evaluation) up to $90/hr. Payout: Payoneer or Tipalti (routes to PayPal).
- [Alignerr: is it legit, what it pays, and live roles](https://aitraining.jobs/platforms/alignerr) - $10–$200/hr; 65 open roles right now. Live median $25/hr across this platform's 65 advertised rates. Top roles: Software Engineer task author $70–$120/hr; Finance & accounting expert $60–$120/hr. Payout: Stripe Connect or PayPal by region; direct deposit.
- [Perle Labs: is it legit, what it pays, and live roles](https://aitraining.jobs/platforms/perle) - Gigs paid per task or audio minute; Points toward PRL in its app; 65 open roles right now. Top roles: Medical text classification (physician review) Points -> $PRL token (higher-tier expert work); RLHF response evaluation Points per task -> $PRL token. Payout: Points redeemable via the $PRL token on Solana (TGE March 2026), plus occasional USDC prize pools ($50-$500) through Galxe quests. Paid on-chain; each task requires an on-chain confirmation (small SOL balance needed).
- [TransPerfect DataForce: is it legit, what it pays, and live roles](https://aitraining.jobs/platforms/transperfect-dataforce) - Per project; ~$10–$31/hr (third-party); 59 open roles right now. Top roles: Data annotation / labeling Per task, varies; AI response rating / relevance Per task or hour. Payout: PayPal, wire, check, Tremendous gift card, or Western Union (by location), after the project quality check clears (~2–3 weeks). W-9/W-8 required.
- [Appen: is it legit, what it pays, and live roles](https://aitraining.jobs/platforms/appen) - $2.50–$14/hr micro; $20–$40/hr evaluators; 58 open roles right now. Live median $30/hr across this platform's 20 advertised rates. Top roles: Search / social media evaluator $20–$40/hr; Bilingual AI language evaluator $7–$25/hr. Payout: Monthly; PayPal/Payoneer.
- [Babel Audio: is it legit, what it pays, and live roles](https://aitraining.jobs/platforms/babel-audio) - $17.50-$150/hr; 43 open roles right now. Live median $105/hr across this platform's 23 advertised rates. Top roles: Voice actor / actress (native language) $60-$150/hr; Transcriber (AI trainer), 20 languages up to $110 per audio hour. Payout: Every Tuesday for the previous week, through Dots to PayPal, Venmo or a bank account; no minimum and no fees.
- [RemoExperts: is it legit, what it pays, and live roles](https://aitraining.jobs/platforms/remoexperts) - $20-$150/hr; 36 open roles right now. Live median $95/hr across this platform's 34 advertised rates. Top roles: STEM expert (chemistry, physics, mathematics) $50-$150/hr; Software / ML engineer (LLM evaluation) $50-$150/hr. Payout: Stripe, Payoneer or Wise; for approved hours, typically monthly.
- [Terac: is it legit, what it pays, and live roles](https://aitraining.jobs/platforms/terac) - $60–$160/hr expert; studies to $300; 20 open roles right now. Live median $75/hr across this platform's 13 advertised rates. Top roles: Finance / domain expert (AI evaluation) $60–$160/hr; Specialist interviews (physicians, professionals) $150–$300/study. Payout: Terac wallet, withdrawable anytime: PayPal, Venmo, Visa prepaid, or bank (US); PayPal worldwide. $10 minimum, $2,000 per-transaction max.
- [Sovrano AI: is it legit, what it pays, and live roles](https://aitraining.jobs/platforms/sovrano-ai) - $35-49/hr expert audits; $150 per accepted deck; $500-1,000 per approved task; 19 open roles right now. Live median $80/hr across this platform's 16 advertised rates. Top roles: Domain expert, AI model audit $35-49/hr; Slides / PPTX expert (task design and rubrics) $150 per accepted deck. Payout: In euros by SEPA transfer or a payout link, into 155 countries (its payout page, 2026-09-01); contractors invoice Sovrano for approved hours or tasks.
- [xAI (SpaceXAI): is it legit, what it pays, and live roles](https://aitraining.jobs/platforms/xai) - $30–$100/hr (US); 17 open roles right now. Live median $40/hr across this platform's 16 advertised rates. Top roles: AI Tutor, language $35–$45/hr (US); STEM tutor (math, physics, chemistry, biology) $45–$75/hr (US). Payout: Hired directly by the company, not through a marketplace wallet: tutor roles are offered as full-time, part-time or contractor depending on the role and the candidate, so pay runs through xAI's own payroll or contract terms. Cadence and method are not published. Eligible US positions include health insurance, a 401(k) and paid sick leave; contractors set their own hours, with most projects needing roughly 10 hours a week.
- [Handshake AI: is it legit, what it pays, and live roles](https://aitraining.jobs/platforms/handshake-ai) - $17–$125/hr; 5 open roles right now. Live median $60/hr across this platform's 5 advertised rates. Top roles: AI Trainer (creative & engineering) $125/hr; Expert AI Trainer (STEM / professional) $60–$125/hr. Payout: 1099 contractor hourly; PayPal, Wise, or bank transfer.
- [Outlier (Scale AI): is it legit, what it pays, and live roles](https://aitraining.jobs/platforms/outlier-scale-ai) - $15–$50/hr; specialists to ~$100/hr; 5 open roles right now. Top roles: Advanced specialist (PhD, legal, medical) $50–$100/hr; Coding / SWE expert $30–$60/hr. Payout: Weekly (Tuesdays) via PayPal, Airtm.
- [Pila8: is it legit, what it pays, and live roles](https://aitraining.jobs/platforms/pila8) - $10-$50/hr (estimated); 3 open roles right now. Live median $30/hr across this platform's 3 advertised rates. Top roles: English conversation recording $20-$30/hr; Single-person script reading $10-$18/hr. Payout: Stripe or Dots after a Persona ID check; withdrawals above a dashboard minimum, every other Wednesday (weekly with the Audio Pro badge).
- [Remotasks: is it legit, what it pays, and live roles](https://aitraining.jobs/platforms/remotasks) - $3–$15/hr; specialist tasks $15–$50/hr; 3 open roles right now. Top roles: Specialist tasks (coding/math/language) $20–$50/hr; RLHF / AI response evaluator $15–$40/hr. Payout: Weekly; PayPal, AirTM.
- [Prolific: is it legit, what it pays, and how to join](https://aitraining.jobs/platforms/prolific) - $8–$150/hr. Top roles: AI Trainer – language fluency ~$30/hr per task; AI Trainer – developer ~$40/hr. Payout: Cash via PayPal only; ~£6/$6 minimum cash-out.
- [DataAnnotation: is it legit, what it pays, and how to join](https://aitraining.jobs/platforms/dataannotation) - $15–$100+/hr. Top roles: Coding / software engineering $50–$100+/hr; STEM expert (math, physics, bio) $50–$100+/hr. Payout: PayPal only; payouts often every few days.
- [User Interviews: is it legit, what it pays, and how to join](https://aitraining.jobs/platforms/user-interviews) - Avg $45+/study; $50–$150+/hr effective. Top roles: Online research study $20–$1,500+/study; 1-on-1 interview $20–$300+/session. Payout: Incentives via Tremendous (virtual Visa, gift cards, or donation); researcher-direct studies may add PayPal/cash/check. Usually within 1–2 days, up to 10 business days.
- [Respondent: is it legit, what it pays, and how to join](https://aitraining.jobs/platforms/respondent) - $50–$400+/study. Top roles: In-person interview / focus group $75–$250+; Specialized professional / executive interview $300–$500+. Payout: Tremendous virtual Visa or gift card only (PayPal retired May 2025); ~7–10 business days after attendance. 5% fee per incentive ($1 minimum).
## Learn each kind of AI-training work (what it is, who it suits, how to start)
- [Learn hub](https://aitraining.jobs/learn) - all 11 kinds of AI-training work grouped into 3 families, each with its barrier to entry and typical hourly range.
- [AI Evaluation & Rating (RLHF): what it is and how to start](https://aitraining.jobs/learn/rlhf-eval) - Rate, rank, and compare AI answers so a model learns which response is better.
- [Coding & Software Evaluation: what it is and how to start](https://aitraining.jobs/learn/coding) - Review, write, and grade real code so AI models learn to program correctly.
- [Math & STEM: what it is and how to start](https://aitraining.jobs/learn/stem-math) - Write and check hard math, physics, chemistry, and biology problems (including formal proofs) that teach and test AI.
- [Law, Medicine & Finance: what it is and how to start](https://aitraining.jobs/learn/domain-expert) - Use your law, medicine, or finance credential to judge whether AI reasoning is actually correct.
- [Data Annotation & Labeling: what it is and how to start](https://aitraining.jobs/learn/data-labeling) - Label images, video, LiDAR, and text so AI models can perceive and categorize the world.
- [Multilingual & Localization: what it is and how to start](https://aitraining.jobs/learn/multilingual) - Give AI models native-language data, translation, and dialect judgment humans cannot fake.
- [Transcription, Audio & Speech: what it is and how to start](https://aitraining.jobs/learn/audio-speech) - Record, transcribe, and label voice audio so AI can understand and produce human speech.
- [Writing & Content: what it is and how to start](https://aitraining.jobs/learn/writing) - Write prompts, model answers, and edits that teach AI to write clearly and well.
- [Red-Teaming & AI Safety: what it is and how to start](https://aitraining.jobs/learn/red-teaming) - Get paid to break AI models: find the unsafe, biased, or wrong answers before the public does.
- [Research Studies & Surveys: what it is and how to start](https://aitraining.jobs/learn/research-studies) - Get paid to take part in surveys, experiments, and human-preference studies that feed AI research.
- [Agentic Tasks & RL Environments: what it is and how to start](https://aitraining.jobs/learn/agentic-eval) - Build and grade the interactive tasks and RL environments that agentic AI trains against.
## In-depth guides
- [AI interviews: how they work and how to pass one](https://aitraining.jobs/learn/ai-interview) - How AI interviews at Mercor, micro1 and Braintrust work, what the AI scores, how to prepare and answer, retake rules, your rights, and what happens after.
- [The fastest-growing jobs in 2026: the top 10, the top 100 and the jobs in decline](https://aitraining.jobs/learn/fastest-growing-jobs-2026) - The 10 job categories adding the most jobs and the most value in 2026, the jobs shrinking fastest worldwide, and the 100 fastest-growing jobs, with numbers.
## Sourcing and honesty
Pay figures are compiled from each platform's own pages, live job feeds, and worker reports (Trustpilot, Glassdoor, Reddit). Live medians reflect open roles as of 2026-10-14; curated benchmarks were last hand-verified 2026-07-25. Each figure carries a confidence label; estimates are marked as such and are never shown as a precise or guaranteed wage.
Sites with a close AI-readiness score — compare how they set up llms.txt and markup.