Handshake AI MOVE Fellowship Review: Is the Domain Specialist Program Worth It? (2026)

Ian Mukanda

28 Feb, 2026

There are two questions every researcher, doctoral student, or subject-matter expert asks when they first hear about the Handshake AI MOVE Fellowship. The first is whether it genuinely pays what it claims. The second is whether the work is actually worth their time — not just financially, but professionally.

This review answers both questions directly, honestly, and with the level of specificity that most review articles on this program deliberately avoid. It covers exactly what the MOVE program is, how it differs from the standard Handshake AI Fellowship, what the work looks like day to day, what fellows actually report about their experience, where the program falls short, and the one factor that will determine whether it is worth it for your specific situation.

No promotional language. No vague impressions. Just a thorough, honest assessment of one of the more significant remote income opportunities available to graduate researchers and domain experts in 2026.

Handshake AI MOVE Fellowship domain specialist program review 2026

What Is the MOVE Program — and How Is It Different From the Standard Fellowship?

The Handshake AI Fellowship has two distinct tracks that are often conflated in online discussions. Understanding the difference between them is the starting point for evaluating whether the MOVE program specifically is the right fit for you.

The Generalist Track is open to associate's and bachelor's degree holders. Tasks are general in nature — evaluating and ranking AI-generated content, assessing multimedia outputs, and applying structured rubrics to broad-scope responses. Pay for generalist work starts around $30 per hour and caps below the domain specialist tier.

The MOVE Track — Model Validation Experts is a fundamentally different program. MOVE stands for Model Validation Experts and is designed exclusively for master's degree holders, PhD candidates, doctoral graduates, and postdoctoral researchers. The work targets the highest-difficulty evaluation tasks that require genuine domain expertise — the kind of knowledge that takes years of focused academic study to develop and cannot be replicated by reading Wikipedia.

The MOVE program offers paid, remote AI work for experts, with fellows earning up to $100 per hour through flexible projects advancing AI models. That positioning is accurate as far as it goes. But the full picture of what that means in practice — including what the work involves, how long it takes to get matched, and what the experience looks like from inside the program — requires more detail than any official page provides.

Related Reading: Handshake AI Fellowship: The Complete Guide to Jobs, Projects, Pay, and Getting Started (2026)

Comparison chart showing Handshake AI Generalist Track versus MOVE Domain Specialist Track

Who the MOVE Fellowship Is Actually Built For

The MOVE Fellowship is designed for masters, PhD, and postdoc students or graduates, giving experts the chance to apply their knowledge to real AI projects. No AI background is required — just deep expertise, curiosity, and the drive to push the boundaries of how knowledge can be shared.

That description captures the eligibility. What it does not capture is the specificity that actually matters for matching. The MOVE program is most valuable — and most likely to result in a fast, well-matched project — for researchers who have genuine depth in a clearly defined subfield, not just broad familiarity with an academic discipline.

A PhD candidate in "biology" describes tens of thousands of people. A PhD candidate in "computational structural biology with a focus on protein-ligand binding dynamics and molecular dynamics simulation" describes a profile that AI labs can match to specific evaluation needs with precision. The more specialized your expertise, the more directly valuable your feedback signal is to the models being trained.

Fields that have historically generated strong MOVE matches include mathematics, physics, chemistry, biology, computer science, linguistics, law, medicine, economics, philosophy, history, and engineering subfields. The program is not limited to STEM — humanities and social science researchers contribute meaningfully to tasks involving historical analysis, legal reasoning, philosophical argumentation, and linguistic accuracy.

In some cases, individuals without a formal degree but with demonstrated subject-matter expertise may be considered. This applies to the MOVE track as well, though it is the exception rather than the standard pathway.

What the MOVE Work Actually Looks Like

This is the section most reviews skip entirely. The official program pages describe tasks in broad categories. What fellows actually spend their working hours doing is more specific — and understanding it is essential for evaluating whether the program suits your working style.

Domain expert researcher evaluating AI model outputs on laptop for MOVE fellowship

Adversarial Prompt Design

The highest-difficulty task category in the MOVE program. You design prompts specifically constructed to expose weaknesses in a model's reasoning within your domain. This is not writing interesting questions — it is engineering inputs that target the exact boundaries of a model's competence in your field.

A mathematician might construct a proof problem that requires recognizing a subtle application of a theorem the model frequently misidentifies. A lawyer might design a fact pattern where the correct legal analysis hinges on a jurisdiction-specific rule that a model trained on general legal text consistently misapplies. A chemist might ask about a reaction mechanism where the correct answer requires distinguishing between two pathways that look superficially similar but proceed through fundamentally different transition states.

This work is demanding and intellectually engaging. It is also the work that pays at the highest tier — because producing genuinely adversarial inputs at the graduate level requires expertise that is irreplaceable.

Expert Response Evaluation

You receive AI-generated responses to domain-specific questions and assess them against a detailed rubric covering accuracy, depth, clarity, completeness, and appropriate level of technical detail. For MOVE-level tasks, this requires more than applying a checklist — it requires actually knowing whether the content is correct, whether it omits critical nuance, and whether it would mislead a student or practitioner who relied on it.

One chemistry domain lead described the experience: "With Handshake, because I'm a chem domain lead, I get to apply my knowledge and explore it in more depth as I help as a reviewer to answer questions and give ideas." That description reflects something genuine about the MOVE experience that distinguishes it from lower-tier annotation work — the tasks require you to engage your expertise at depth, not just apply a surface-level rubric.

Comparative Ranking With Structured Justification

You are presented with two or more AI-generated responses to the same domain-specific prompt and required to rank them with detailed, structured written justifications. On MOVE projects, these justifications are evaluated as carefully as the rankings themselves — reviewers assess whether your reasoning is specific, accurate, and correctly applies the rubric criteria.

The quality of your written justifications directly determines your performance score on the project, which in turn influences your access to future work and higher-tier assignments.

Domain Lead Responsibilities (Senior MOVE Fellows)

Some fellows who demonstrate consistent quality and deep expertise are offered domain lead roles within their project. Domain leads take on additional responsibilities — reviewing other fellows' work, providing feedback, answering domain-specific questions from the broader team, and mentoring newer contributors.

Domain lead roles come with elevated compensation and more varied work. They are not part of the initial application — they are earned through demonstrated performance on standard MOVE tasks.

The Pay Reality: What Fellows Actually Earn

Let's be direct about compensation, because the "$100 per hour" headline is accurate in context but requires qualification to be useful.

Pay rate breakdown chart for Handshake AI MOVE Fellowship domain specialist levels

Standard MOVE rates by credential level:

Master's degree contributors: approximately $50–$75 per hour depending on project and domain
PhD candidates and graduates: approximately $75–$100 per hour on standard MOVE projects
Domain leads and senior specialist roles: $100 per hour and above on select projects

These rates are transparent — every project publishes its hourly rate before you accept it, and rates do not change mid-project. As one fellow noted: "I feel like our needs as fellows are being considered very thoughtfully. And I think it's really well organized. The pay is awesome."

What is equally important to understand is what drives earnings in practice — because hourly rate and actual earnings are related but not identical.

Factors that affect real weekly earnings:

Project availability. MOVE projects are launched based on partner lab needs. There are periods of high task availability and periods where task queues are lighter. Fellows cannot control project launch timing. During high-availability periods, earning $1,500–$2,500 per week at 20 hours is realistic for PhD-level contributors. During lighter periods, weekly earnings may be significantly lower regardless of your hourly rate.

Task acceptance timing. Tasks are available in queues and are claimed on a first-come basis in some projects. Fellows who log in regularly and respond quickly to task availability tend to secure more hours than those checking infrequently.

Quality score. Your performance on quality assessments directly affects the volume and tier of tasks offered to you. High-scoring fellows receive more tasks, more consistently. Maintaining accuracy is not just a program requirement — it is an earnings multiplier.

Onboarding period. The onboarding phase includes approximately two hours of paid learning modules, a knowledge quiz, and a live practice task with reviewer feedback. This time is compensated, but the first one to two weeks of any project involve a ramp-up period where task volume is lower than it will be once you are fully embedded.

Related Reading: Highest Paying AI and LLM Training Jobs for Students and Researchers in 2026

The Application and Matching Process: What to Realistically Expect

Handshake AI MOVE Fellowship application process timeline step by step

The MOVE application process is more rigorous than most candidates expect, and the matching timeline is longer than most candidates want. Both facts are worth understanding clearly before you begin.

Applications are reviewed on a rolling basis according to partner lab needs, and candidates are selected based on their academic background and domain knowledge. Selection timelines vary — just because a candidate has not heard back does not mean they were not selected. When there's a project that matches their expertise, the team will reach out with an invitation via email.

In practical terms, this means the following:

Assessment quality is the first filter. The initial assessment is not a formality. It tests both your domain knowledge and your ability to follow detailed evaluation instructions — with equal weight on both dimensions. Strong domain experts who read instructions carelessly fail the assessment. Students with moderate domain expertise who apply the rubric precisely often advance when more credentialed candidates do not.

Matching timelines are genuinely variable. MOVE candidates with highly specialized expertise in domains currently aligned with active lab projects can be matched within two to four weeks. Candidates in domains with lower current project demand may wait several months. The wait does not indicate rejection — it reflects the rolling project launch schedule.

Completing the onboarding checklist immediately is critical. Once signed up, fellows are matched to relevant projects based on their experience and background, with projects available on a rolling basis. An incomplete onboarding checklist — missing tax forms, identity verification, or work authorization documents — removes you from consideration for matching until it is complete. Do not leave any step unfinished.

Your profile specificity is the primary matching variable. The single most impactful action you can take to accelerate matching is writing your expertise description in specific, subfield-level terms. "PhD in Chemistry" does not match. "PhD candidate in physical organic chemistry, dissertation on solvent effects on reaction rates in polar aprotic media, with additional expertise in NMR spectroscopy and computational DFT modeling" matches to chemistry-specific projects with precision.

Related Reading: How to Write a Strong Handshake AI Fellowship Application Profile That Gets Matched Faster

What Fellows Actually Say: Real Voices From Inside the Program

Handshake AI MOVE Fellow testimonials showing real experiences from domain specialists

The most useful signal for evaluating any program is what current participants say when they are not being prompted to be promotional. Here is what fellows across different academic backgrounds have reported about their MOVE experience.

For Nasser Heyradi, PhD in Mathematics, the fellowship felt like the right fit — where both math skills and life experience could help AI learn in new ways. "After three or four months of working with Handshake, I had this feeling, 'This is the right place for me.' I can use not only my math skills here, but also my life skills — because we want to make the model learn something new."

For Kat Mueller, PhD in Cell and Molecular Biology, the fellowship stood out because it felt thoughtfully designed and offered work she never expected to find outside the lab.

For Tosha Laughin, PhD in Chemistry who became a domain lead, the program created space to mentor others while deepening her own expertise.

Another fellow reflected: "Before working with Handshake AI, I thought that it could never be an opportunity to work in tech because of my background. Now I feel fully confident that I can work in tech. That it doesn't matter what background I have and this is an amazing opportunity to apply."

A separate fellow observed: "You get to see how civil engineers, teachers, and web developers would do their work from day to day. So I've knocked the rust off of my old skills that I used to use pretty regularly, but then also I get to see some new ones, and that's kind of enlightening."

These testimonials reflect a consistent pattern across the MOVE experience: fellows find the work more intellectually engaging than they expected, the pay consistently meets or exceeds what was advertised, and the program's structure — its organization, transparency, and task quality — stands out positively relative to other AI training platforms.

The Honest Limitations: Where the MOVE Program Falls Short

A review that only covers what works is not a review — it is an advertisement. Here are the genuine limitations of the MOVE program that are worth knowing before you commit time to the application process.

Weighing pros and cons of Handshake AI MOVE Fellowship for graduate researchers

Project availability is not guaranteed or consistent. The MOVE program does not offer a guaranteed minimum number of hours per week. Project availability fluctuates based on partner lab project cycles. There will be high-volume periods and quieter periods. Fellows who depend on a predictable income stream may find this frustrating. The program is most sustainable as supplemental income alongside a primary role such as a research stipend, rather than as a standalone primary income source.

The matching wait can be significant. For candidates in fields with lower current project demand, the wait for a first project can extend to several months. This is not a program flaw — it reflects the reality that matching is based on lab needs, not candidate timelines. But it means the application process rewards patience rather than urgency.

The program is U.S.-only. Participation requires being physically based in the United States with valid work authorization. International researchers who are not U.S.-based or do not have qualifying work authorization cannot participate, regardless of their expertise level.

STEM OPT with I-983 is not supported. F-1 students on STEM OPT who need an I-983 training plan from their employer are not eligible. This is a meaningful gap that affects a specific but significant population of international graduate students.

The confidentiality constraint limits specific attribution. Fellows are encouraged to add their Handshake AI Fellowship experience to their LinkedIn, Handshake profile, or resume and CV, but should not include the name of the partner company or AI lab, as these projects are confidential. This limits the specificity with which you can describe the work on professional profiles — a consideration for researchers who want to cite specific lab relationships in their career narrative.

Domain lead roles are earned, not assigned. If you are hoping to begin immediately in a leadership or senior review capacity, the program does not work that way. Domain leads emerge from demonstrated performance over time. Entry to the program is always at the standard contributor level.

How MOVE Fellowship Experience Translates to Your Professional Profile

One of the most underappreciated aspects of the MOVE program is what it produces beyond income. For researchers navigating the academic-to-industry transition, or for graduate students building a professional identity outside their university lab, the MOVE experience is genuinely marketable.

Suggested language for professional profiles, adapted from Handshake's own guidance: "Developed and evaluated domain-specific prompts to assess the performance of large language models in [domain]," "Analyzed LLM outputs for scientific accuracy, clarity, and depth in specialized subfields," "Contributed to improving AI understanding of complex topics through expert review and feedback," and "Conducted independent research to support prompt development and evaluation tasks."

Customize each description to reflect your actual domain and task types. AI evaluation experience at the domain specialist level is increasingly recognized by technical employers, AI policy organizations, and research institutions as meaningful AI/ML exposure — not simply as annotation work.

For PhD students considering careers in AI policy, AI safety, or AI ethics, MOVE experience provides direct familiarity with how model evaluation works in practice — a perspective that most professionals in those fields develop only theoretically.

Related Reading: AI Annotation Jobs Explained: What Tasks You Do, How Much You Earn, and Which Platforms Are Worth It

MOVE vs. Competing Domain Specialist Programs: How It Compares

Comparison table of Handshake AI MOVE fellowship versus Scale AI and DataAnnotation for domain experts

The MOVE program does not exist in isolation. Domain specialist AI training work is available through several other platforms in 2026. Here is how MOVE compares directly to the most relevant alternatives.

Handshake AI MOVE vs. Scale AI Expert Contributor Program

Scale AI's expert contributor track recruits domain specialists across technical and professional fields and pays $40–$80 per hour for specialist evaluation work. The Scale platform offers broader task variety and more consistent task availability than MOVE, but the application process is similarly selective. Scale's payout is weekly via Stripe or PayPal. Tax documentation is less automated than Deel-based 1099 generation. For technical contributors — engineers, developers, data scientists — Scale's expert track is a strong alternative. For academic researchers in non-technical fields, MOVE's domain matching is typically more precise.

Handshake AI MOVE vs. DataAnnotation.tech Expert Projects

DataAnnotation.tech offers expert-level projects at $40+ per hour for credentialed contributors. The platform is more accessible than MOVE in terms of onboarding speed — contributors can often begin earning within two weeks. Pay ceiling is lower than MOVE's top tier, and the task types are weighted more toward coding and writing evaluation than toward discipline-specific academic reasoning. For researchers who want faster onboarding at a slightly lower pay rate, DataAnnotation is a reasonable complement to a MOVE application.

Handshake AI MOVE vs. Prolific Research Studies

Prolific recruits academic researchers for paid research participation at $15–$50 per hour for AI training studies. The work is fundamentally different — participation in research studies rather than systematic model evaluation. Prolific is global, requires no application process, and is accessible immediately. It is not a direct competitor to MOVE but is a worthwhile supplemental platform for researchers waiting for their MOVE project match to materialize.

Related Reading: Deel vs Stripe for Freelancers in 2026: Which Platform Pays Faster and Has Lower Fees?

The Verdict: Is the Handshake AI MOVE Fellowship Worth It?

The answer depends on two things: your credential level and your income expectations.

It is clearly worth it if:

You hold a master's or PhD in any academic field, are U.S.-based with valid work authorization, can invest the time to write a precise and specific profile, are willing to complete the assessment carefully, have the patience for a matching process that may take several weeks, and are treating the income as supplemental to a primary academic role rather than as a standalone salary replacement.

In that scenario, the MOVE fellowship offers the highest accessible hourly rate for remote part-time work that directly leverages your academic expertise — up to $100 per hour — with genuine schedule flexibility, paid onboarding, clean tax documentation through Deel, and work that is intellectually meaningful rather than rote.

It may not be worth the time investment if:

You need guaranteed, predictable weekly income immediately. You are on STEM OPT requiring an I-983. You are located outside the United States. You have low tolerance for variable task availability or an extended matching wait. Or you are a bachelor's-level contributor who would be better served by the generalist track or by platforms with faster onboarding at the mid-pay tier.

The MOVE fellowship is a premium program that rewards patient, credentialed applicants who engage with it seriously. For that audience, it delivers on its promises — and delivers well.

Related Reading: Best Remote Part-Time Jobs for Graduate Students in 2026 That Pay Over $50 an Hour

Frequently Asked Questions

What does MOVE stand for in the Handshake AI MOVE Fellowship? MOVE stands for Model Validation Experts. It is the domain specialist track of the Handshake AI Fellowship, designed for master's, PhD, and postdoctoral researchers who evaluate and improve AI models in their specific academic field.

How much does the Handshake AI MOVE Fellowship pay per hour? Domain specialist fellows earn $50–$75 per hour at the master's level and $75–$100+ per hour at the PhD and domain lead level. Rates are published before project acceptance and never change mid-project.

How long does it take to get matched to a MOVE project? Matching timelines vary from two to four weeks for candidates whose expertise aligns with active projects, to several months for those in less currently active domains. Completing your onboarding checklist immediately and writing a highly specific expertise description are the two most effective ways to reduce wait time.

Can I do MOVE fellowship work alongside my PhD program? Yes. The program is designed specifically for active graduate students. Work is fully asynchronous, part-time, and capped at 40 hours per week maximum. Most MOVE fellows contribute 10–20 hours per week alongside active research.

Does MOVE work count toward CPT or OPT authorization? CPT and standard OPT are accepted. STEM OPT requiring an I-983 is not supported. Confirm eligibility with your Designated School Official before accepting any project contract.

Can MOVE fellowship experience be added to a resume? Yes. Fellows are encouraged to add their experience to LinkedIn and resume profiles using domain-specific language. The name of the partner AI lab cannot be disclosed, but the nature of the work — domain expert LLM evaluation, adversarial prompt design, expert response assessment — can and should be represented accurately.

What happens if my quality score drops on a MOVE project? Low quality scores reduce task allocation and can ultimately result in removal from a project. Quality is assessed through inter-annotator agreement checks and rubric adherence audits. Maintaining quality requires reading guidelines thoroughly before beginning each task type, not just at initial onboarding.

Handshake AI Fellowship: The Complete Guide to Jobs, Projects, Pay, and Getting Started (2026) — the complete guide to the broader Handshake AI program, including payment methods, eligibility, and payout schedules.
How to Write a Strong Handshake AI Fellowship Application Profile That Gets Matched Faster — section-by-section guide to writing the profile that gets your MOVE application matched faster.
Highest Paying AI and LLM Training Jobs for Students and Researchers in 2026 — where the MOVE fellowship sits in the broader landscape of high-paying AI training roles.

External Resources

Handshake AI Fellowship Official Page: joinhandshake.com/fellowship-program — apply directly and review the official program overview.
Handshake AI MOVE Program Page: joinhandshake.com/move-program — the dedicated MOVE domain specialist landing page with application link.
Handshake AI Fellow Stories: joinhandshake.com/blog/students/handshake-ai-fellows — real testimonials from current MOVE and generalist fellows across academic disciplines.
Handshake AI Help Center: support.joinhandshake.com — official documentation on onboarding, payment, eligibility, and program guidelines.
IRS Self-Employed Tax Center: irs.gov/businesses/small-businesses-self-employed/self-employed-individuals-tax-center — all MOVE income is independent contractor income. Understand your tax obligations before your first payout.

Disclosure: This article is independently researched and is not sponsored by or affiliated with Handshake. Fellow testimonials are sourced from publicly available Handshake blog content and official program pages. Pay rates, eligibility criteria, and program details are subject to change. Always verify current information directly through Handshake's official Help Center before applying.