# Mercor Jobs in 2026: What Students Should Actually Know **Updated July 2026** ## Quick answer Mercor is a legitimate expert marketplace that matches vetted professionals to AI labs and pays reliably every week. But it's expert-gated: most roles want 3+ years of experience or an advanced degree, so fresh undergrads rarely get matched. Grad students and genuine specialists do. Beginners should build eval experience elsewhere first. ## Why you're probably here You saw a video. Someone on TikTok held up a Stripe screenshot, said "$150 an hour training AI," and named Mercor. So you searched it. Here's the honest version before you spend an evening on an application. Mercor is real. It pays on time. And it is very likely not for you yet — not because you're not smart enough, but because of who it's built to hire. This guide is the expectation-setter: what Mercor actually is, what it really pays, how the process works, who gets in, and the specific path a student can take to get there in a year or two instead of getting a silent rejection this week. If you want the broad map of beginner-accessible AI work first, that lives in the [entry-level AI jobs](/entry-level-ai-jobs/) hub — and notice Mercor isn't in its beginner table. That's deliberate. It's expert-gated, and this page explains why. ## What Mercor actually is Most of the platforms students start on — DataAnnotation, Outlier, Clickworker — hire crowds to do task work. You sign up, pass an assessment, and claim tasks from a queue. Mercor is a different model. It's a **talent marketplace**: it matches individually vetted, credentialed professionals to specific contract work at AI labs. The clients matter here, because they explain the bar. Mercor reportedly places people on projects for AI labs including OpenAI, xAI, and Scale. When your customer is a frontier lab paying for expert judgment — a physician labeling medical reasoning, a senior engineer grading code, a lawyer reviewing legal outputs — you don't want a crowd. You want a small, verified pool of people who can prove they know the domain. That's the whole design. Mercor's value to its clients *is* the filter. Which means the filter is aimed squarely at people with real credentials, and pointed away from beginners. So "Mercor jobs" aren't a queue you dip into for beer money. They're contract placements, matched to you, based on a profile the platform has vetted and scored. ## The pay reality *Ranges compiled from platform listings and worker reports · last verified July 2026.* Here's the part the videos get wrong, laid out as claims versus what workers actually report. **What Mercor's marketing claims:** expert roles at **$50–$150/hour**, with occasional references to **$200/hour** for elite specialists. Those numbers are not fake. They're real rates paid to real people at the top of the pool. **What workers actually report earning:** - **General evaluation work: about $20–$30/hour.** This is the tier closest to what a strong student might reach. - **Software / coding work: about $35–$60/hour.** - **Domain experts: $40–$70+/hour**, with individual anecdotes of $150/hour side gigs. Read those two lists together and the picture is clear. The $150/hour figure is a ceiling for credentialed specialists, not a starting rate. The general-eval tier — the realistic entry point — pays roughly what a good AI-training platform pays, without the multi-week wait to get started. The premium exists, but it's gated behind exactly the expertise a first-year undergrad hasn't built yet. That's not a knock on Mercor; it's the shape of every expert marketplace. ## The process, step by step Getting into Mercor is a screening, not a signup. It runs in three stages: 1. **Profile and resume.** You build out your background — degrees, work history, domain skills. This is where the gate first bites: a thin resume with no experience and no advanced coursework rarely clears it. 2. **A one-way AI video interview.** You record answers to prompts; there's no live human on the other end. The interview is scored, and that score is a major input into whether — and where — you get matched. Treat it like an oral exam in your field, not a casual chat. It's measuring whether you can reason clearly and demonstrate real depth in the domain you claimed. 3. **The "Verified Expert" pool.** Clear the bar and you're added to a vetted pool, then matched to projects as they come up. Expect a **wait of two to four weeks (often more)** before your first project lands. That wait is the single most common complaint about the platform — and notably, it's a complaint about *matching speed*, not about getting paid. ## Who actually gets in Blunt version: Mercor is the **weakest fit on the whole AI-work landscape for a fresh undergraduate.** Most roles want three or more years of professional experience or an advanced degree. If you're a sophomore with no work history, the profile stage will likely screen you out, and that's the system working as designed. Who does get in: - **Grad students** — a master's or PhD in progress is real credentialing, especially in STEM, medicine, or law. - **Specialists with deep domain skill** — you've worked as a nurse, a developer, an accountant, a translator, and you're now a student. Your prior career is the credential. - **Strong technical undergrads, sometimes, via the general-eval tier.** If you're a junior or senior in CS, math, or a pre-med track with genuine, demonstrable ability, the general-evaluation tier (~$20–$30/hour) is the realistic door — not the $150/hour expert placements. Go in expecting that tier, and you won't be disappointed. If that's not you yet, don't take it personally and don't fake credentials to slip through — the AI interview is built to catch exactly that. Build the track record first. ## The student path *into* Mercor Here's the useful part, because "not yet" isn't "never." Mercor rewards a demonstrated history of doing AI evaluation and domain work well. You can build exactly that history on platforms that *will* take a beginner — and then come back to Mercor with something the profile stage actually recognizes. The on-ramp: - **Start where the door is open.** Build real evaluation and RLHF experience on beginner-friendly platforms — the ones covered in [AI training jobs](/ai-training-jobs/). DataAnnotation and Outlier will take a strong student without years of experience, and the work is the same *kind* of work Mercor's clients pay for: grading model outputs, writing reference answers, catching errors. - **If you have subject depth, lean into it.** A technical major is a credential in disguise. The specialist tier — tutoring AI models in your field — is the closest thing to Mercor's expert work that a student can reach today, and it's covered in [AI tutor jobs](/ai-tutor-jobs/). That guide already flags Mercor as the top of the pay ladder and the weakest fit for undergrads; think of the tutor platforms as the training ground for it. - **Then come back.** After six to twelve months of paid eval work, a semester or two further into your degree, or with a graduate program underway, your Mercor profile reads completely differently. Same platform, same interview — different candidate. That's the honest route the videos skip. Mercor isn't a first job. It's a good second or third one. ## A warning: fake Mercor recruiters Because Mercor's name carries weight now, scammers borrow it. Fake "Mercor recruiters" reach out over WhatsApp, Telegram, or text, offer you a spot, and then steer you toward something a real employer never does — a fee, a check to deposit and partly wire back, or a request for bank logins. Mercor's actual process starts with *you* applying on its official site; a legitimate placement never begins with an unsolicited DM promising you a role. Apply only at Mercor's own domain, and if a "recruiter" contacts you first, treat it as a scam until proven otherwise. The full checklist for telling a real platform from an impersonator is in [is data annotation legit](/is-data-annotation-legit/). ## Tools that get the interview Knowing your subject is what clears Mercor's AI interview — no tool substitutes for that. But when you're applying across Mercor and the on-ramp platforms at once, or building the resume that finally clears the profile gate, a few tools save real time. Our current picks — with the honest caveats and what each actually costs — live on one page: **[the tools we actually recommend](/tools/)**. ## FAQ **Is Mercor legit?** Yes. Mercor is a real expert marketplace that matches vetted professionals to AI labs, and it pays reliably — weekly via Stripe. The recurring complaints are about project cancellations and the wait to get matched, not about non-payment. The one real risk to watch is impersonation: fake "Mercor recruiters" who contact you first are scams, not the company. **Does Mercor really pay $150/hour?** For some people, yes — that rate is real for credentialed specialists at the top of the pool. But it isn't a starting rate. Worker reports put general evaluation work around $20–$30/hour, software work around $35–$60, and domain experts at $40–$70+. Treat "$150/hour" as the ceiling for deep expertise, not the number you'll see first. **Can students get on Mercor?** Rarely as fresh undergrads — most roles want three or more years of experience or an advanced degree. Grad students and people with a prior professional career get in more easily. A strong technical junior or senior might reach the general-evaluation tier (~$20–$30/hour), but not the expert placements. Most students should build experience elsewhere first. **How long until your first Mercor project?** Plan for two to four weeks, often longer, between clearing the vetting and getting matched to your first project. This matching wait is the platform's most common complaint. It's a scheduling reality of the marketplace model, not a payment problem — once you're working, pay arrives on time. **What's the Mercor AI interview like?** It's a one-way, recorded video interview — you answer prompts on your own, with no live interviewer. It's scored, and that score heavily influences whether and where you get matched, so it's a major pay determinant. Prepare for it like an oral exam in your field: it's designed to test real depth, and it's built to catch people bluffing credentials they don't have.