✅ verified listings — click any card to see full details & apply
Surge AI trains frontier models for OpenAI, Anthropic, Meta and Google — and they hire the world's actual elite: board-certified MDs ($250–450/hr), Supreme Court lawyers ($500–1,000/hr), McKinsey consultants ($300–500/hr), Goldman Sachs bankers ($300–500/hr), NYT journalists ($200–400/hr), VC partners and startup CEOs ($500–1,000/hr), and university professors ($150–300/hr). Apply by emailing [email protected] — not through a form.
Tether has 250 open roles spanning AI Research Engineering (LLM, Multi-Modal, Vision), Backend Development, ML Engineering internships, Marketing, Finance and Operations — with India-specific roles in Bangalore and Gurgaon. Apply only through the official careers page. All legitimate recruiters contact via @tether.to or @tether.io emails only — never WhatsApp or Telegram.
ElevenLabs is the world's leading AI voice platform ($3.3B valuation), offering 150+ open roles across engineering, growth, revenue, and operations — plus separate freelance roles for audiobook specialists, dubbing specialists, translators, and audio engineers. India-specific roles are available. This is a rare platform offering both full-time careers AND paid freelance work at the same company.
Scholars by iMerit is an expert network for PhDs, MDs, linguists and engineers who do AI training and chain-of-thought reasoning work for LLM fine-tuning. Unlike anonymous gig platforms, iMerit puts you on a small team with a project manager rather than treating you as part of a crowd — backed by 12+ years of company history.
AI Acquisition is a marketing/growth agency that sells AI-powered business systems to entrepreneurs — NOT an "earn from AI" platform. Their careers page lists 50+ genuine internal job openings (Marketing, Operations, Customer Success, Finance) across 20+ countries including India. Treat this as a normal remote job application, not a gig-economy opportunity.
Snorkel AI offers two genuinely different paths: full-time corporate roles in Engineering, Research, Sales and Product (mostly based in San Francisco/Redwood City, hybrid), and a separate Expert Community for PhDs and domain specialists who want contract-based, remote-friendly work building AI training datasets. Pick whichever matches what you're looking for.
Avalara is a publicly-traded (NYSE: AVLR) tax compliance and AI platform with 200,000+ customers across 75+ countries. This is a stable, traditional corporate job rather than a freelance gig — Glassdoor shows a 3.2/5 employee rating with 49% recommending it to a friend, which is solid but not exceptional. Open roles include account management, benefits systems, and engineering positions, with some based in Pune, India.
Besimple AI is a Y Combinator-backed startup building the data layer for voice AI. Their contributor side, Pila8, pays everyday people (not just engineers) to record short voice clips that train AI to understand different accents and languages — base rate is $0.10/minute, with non-English speakers and consistent contributors earning bonuses.
DataForce is TransPerfect's AI data platform with over 1.3 million remote contributors worldwide, working on data annotation, transcription, search evaluation, and AI training tasks. It's a legitimate, established side of TransPerfect (25+ years, 140+ cities) — but reviews say to treat it as flexible freelance income, not a stable full-time job, since project availability varies by location and language demand.
Revolut is a global fintech "super app" with 60+ million customers and 622 open roles across Engineering, Compliance, Sales, Finance, and Legal. It's genuinely remote-first — 239 of its open roles are fully remote — and known for fast growth, strong pay (engineers average $126K+), and an intense, results-driven culture rather than a relaxed one.
Welo Global is a 25+ year old language and localization company trusted by Google, Spotify, Amazon, and other major brands — they handle translation, transcription, data annotation, and AI training tasks across 300+ languages. It's one of the more established names in this space, which generally means more structured onboarding and steadier project flow compared to smaller gig platforms.
Crossover has run remote, performance-based hiring for years, placing people in roles with Fortune 500 clients across 100 countries, with a wide pay band reflecting very different role levels. The "top 1%" framing means the screening process is intentionally selective.
GitLab is a major, publicly traded (Nasdaq: GTLB) DevOps platform company that's run fully remote since it was founded — no offices, no headquarters, just a globally distributed team across 65+ countries. It's a different kind of listing than most others here: this is full-time employment at an established public tech company, not freelance or gig-based AI training work.
Built on Handshake's existing university network, this fellowship targets PhD and Master's-level fellows who evaluate or generate advanced AI responses in their academic specialty. It's one of the higher-paying options here, but it leans heavily on formal qualifications.
Pareto AI runs a human-data platform that connects AI labs with vetted subject-matter experts for RLHF, prompt evaluation, and domain-specific annotation — across fields like medicine, law, finance, and engineering. It sits at the high end of this list: the company says general work pays $35–$60/hr, with specialized expert tasks going up to $230/hr, and that it has raised $37M in funding so far.
Rex.zone connects subject-matter experts with AI labs that need domain-specific data — model answers get checked, scored, or rewritten across fields like law, medicine, finance and more. Work is task-based and remote, so you pick projects that match what you actually know.
Scale AI is one of the most established names in AI training data, founded in 2016 and used by many major AI labs. As the most well-known platform on this list, expect higher applicant volume and correspondingly more competitive task qualification.
Stellar AI is a freelance AI training platform open to contributors worldwide with no prior experience required — you complete tasks like rating responses, writing prompts, or evaluating AI outputs, and get paid hourly. The "$25/hr+" framing means that's the starting rate, with higher-paying tasks available depending on skill and task type. Being "open-ended contract" means there's no fixed end date, but also no guaranteed hours — work is available when projects are live.
Alignerr recruits ML experts to label and validate robotics-related training data. Worth flagging clearly: the $1000 shown on this card is a referral bonus for bringing in other qualified experts, not the hourly pay for the work itself — actual task pay is set separately once you're onboarded.
AfterQuery is a YC-backed startup hiring AI experts to evaluate and refine model outputs across 40+ open roles. Being YC-backed means it went through outside investor vetting, which is a decent — though not guaranteed — signal of legitimacy.
One of the more established platforms on this list — contributors complete coding, writing, and data-labeling tasks that help train AI models, paid hourly. Getting in usually means passing a qualification exercise first; pay tends to rise as you build a track record.
An invite-only network for top-tier AI experts. Invite-only models usually mean a higher entry bar and limited slots, so this one's less about applying on demand and more about getting noticed by the network in the first place.
Innodata is a publicly traded (Nasdaq) AI data engineering company that's been operating for decades, working with major tech clients. As a listed public company, it has more public financial transparency than most platforms on this page.
Focused specifically on language and audio work — transcription, voice recording, pronunciation review for AI speech models. It's a narrower niche than most listings here, so it suits people with strong audio or language skills more than general data work.
Mercor matches experts directly with frontier AI labs for premium-paying contract work, with pay varying a lot by skill level and project. It's one of the higher-end options on this list, which usually also means more competitive screening.
Similar in model to Mercor and Turing — micro1 matches AI experts with frontier lab projects, with pay scaling by skill level. A few platforms on this page use this same "staffing agency for AI labs" approach, so requirements tend to overlap.
Mindrift runs AI training projects across many skill areas and languages, with pay scaling from entry-level to expert tasks. It's a broad, multi-skill platform rather than one narrow niche, so most backgrounds can usually find something to apply for.
Listed as full-time remote tech roles rather than freelance gig work, with 25+ positions open. Full-time listings are worth a closer look at company background, since the commitment level here is higher than a freelance task platform.
OneForma focuses on linguistic and data-collection microtasks — translation checks, transcription, search relevance — across 60+ languages. Pay per task is usually modest since tasks are short, but it's accessible if your language isn't widely supported elsewhere.
Outlier runs expert evaluation tasks — reviewing, scoring, and correcting AI-generated answers — across 60+ open roles in subjects like coding, math, and writing. Pay is task-based rather than a flat hourly rate, so it depends on which project you qualify for.
Prolific is a bit different from the rest of this list — instead of AI-labeling tasks, it connects you with paid academic and commercial research studies and surveys, some of which feed into AI/UX research. Pay is per-study rather than ongoing hourly task work.
RWS is an established language-services company expanding into AI training data work through TrainAI, covering translation-adjacent and linguistic annotation tasks. Pay here sits on the lower end of this list, which is typical for high-volume linguistic microtasks.
A part-time Prompt Engineer role — writing and refining prompts used to test or steer AI model behavior. It's narrower and more specialized than most listings here, and the pay reflects a part-time rather than full freelance commitment.
TELUS Digital's AI data arm runs annotation and labeling projects worldwide, currently with seven open contributor roles. Being part of a large, established company, the process tends to be more structured than smaller gig platforms.
Turing places software and AI engineers with major companies on a project basis, including LLM-training work like agent function-calling — teaching a model to correctly trigger tools and APIs. Engagements are typically several weeks long rather than one-off tasks.
Listed as a flexible freelance contributor role. We don't have much first-hand detail on this one yet, so treat it like any unfamiliar gig listing: read the actual offer page closely before committing real time to it.
xAI is running a large recruiting push — 219 listed roles — for people who help train and evaluate Grok. A drive at this scale usually means varied roles at very different pay levels, so "up to $200/hr" applies to specific senior tasks, not every opening.
Appen is a long-running, publicly listed data company, and CrowdGen is its crowdsourced worker portal for things like search evaluation, data collection, and transcription in 200+ countries. It's a high-volume microtask model — individual tasks pay modestly but availability is broad.
SuperAnnotate's subject-matter-expert track pays a wide range depending on how specialized your knowledge is, with roles open across 40+ countries. The wide $20–$130+/hr band reflects how much subject expertise can change your rate.
Invisible Technologies runs the Meridial Expert Network, an operations-and-data partner for several major AI labs, with a large existing pool of experts already onboarded. Being part of a bigger network can mean steadier project flow, but also more competition to get matched.
HowToAIjob is a listings directory for the AI-economy job market — the fast-growing world of AI training, data annotation, model evaluation, and expert-review work that companies like Scale AI, Mercor, and Outlier rely on to build better AI systems. Instead of checking a dozen different company career pages every day, you can scan one page here and see what's currently open, roughly what it pays, and whether it's freelance or full-time.
We don't run any of these jobs ourselves, and we're not affiliated with the companies listed below. We track public hiring pages across the AI training industry and pull the details into one place, updated as listings change. When you click through on a card, you're applying directly with that company, on their own site, under their own terms — we're simply the map, not the destination.
We scan the industry
Our team checks public career and hiring pages across dozens of AI companies and platforms for new openings in training, annotation, evaluation, and related freelance or full-time roles.
We sort and tag
Each listing gets sorted by type — freelance, full-time, or part-time — tagged with pay range and location where available, and marked "New" when it's been posted recently.
We add real context
Tap any listing and you'll find a plain-language summary: what the work actually involves, who it tends to suit, and what's worth knowing before you apply — not just a one-line title.
You apply, directly
"Visit official page" sends you straight to the company's own application page. We don't collect applications, charge fees, or sit anywhere in the middle of the actual hiring process.
Are these AI jobs legit?
Yes — every listing links straight to the company's own official careers or hiring page, and we note only public, verifiable openings. We're not affiliated with any of these companies and don't collect applications ourselves, so you always apply directly on their site under their own terms. Always confirm details before starting work or sharing personal information, and never pay anyone to "secure" a job.
Do I need experience for AI training and data annotation jobs?
Most data annotation, RLHF, and AI training roles are entry-friendly and don't require a technical background — companies like Outlier, DataAnnotation.tech, and Prolific mainly look for strong written English and attention to detail. Some specialist and expert-review roles (medical, legal, coding) do require relevant credentials or domain expertise, which is noted on each listing.
How much can I earn from AI jobs and data annotation work?
Pay varies widely by platform and specialization — general annotation and evaluation work typically pays in the $15–$50/hr range, while credentialed expert reviewers (doctors, lawyers, senior engineers) can earn $100–$1,000/hr on platforms like Surge AI. Each listing on this page shows its own typical pay range.
Is HowToAIJob free to use?
Yes, completely free. We're an independent listings directory, not a recruiter — we don't charge job seekers, and we don't sit between you and the hiring company. Browsing, filtering, and applying through any listing costs nothing.
Can I do these AI jobs remotely from India or other countries?
Most listings here are 100% remote and open to applicants in many countries, including India — several companies (Tether, ElevenLabs, DataForce) specifically call out India-based roles. Each card notes location requirements where the company restricts them.
How often are new AI jobs added?
We check public hiring pages across the AI training industry regularly and update listings as they change — new or recently updated roles are marked "New" on their card, and the "Last updated" date at the top of the page reflects the most recent refresh.
Don't miss the next opportunity
Join 10,000+ people getting daily AI job alerts — completely free.