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

Ian Mukanda

17 Feb, 2026

If you're a student, recent graduate, or subject-matter expert looking for legitimate remote work that pays well, the Handshake AI Fellowship may be one of the most overlooked opportunities of 2026.

Unlike typical gig platforms that pay a few dollars per task, this program connects qualified individuals directly with frontier AI labs — and the hourly rates reflect that. Before you apply, it helps to understand exactly what the work involves, who can participate, how payments work, and what to realistically expect. This guide answers every question you're likely to have.

What Is the Handshake AI Fellowship?

Handshake — the career platform trusted by over 18 million students across 1,600+ higher education institutions — launched the Handshake AI Fellowship as a dedicated initiative for training and evaluating large language models (LLMs).

Handshake AI Fellow Program | Feedkin.com

The program connects current students, recent graduates, and professionals with flexible, project-based opportunities to help improve the AI models developed by leading labs. Fellows contribute their human judgment and subject-matter expertise to strengthen model performance in specific academic and professional domains.

Think of it this way: frontier AI labs need humans with real academic depth to design challenging prompts, evaluate model outputs, and identify where an AI's reasoning breaks down. A chemistry PhD catches errors a generalist would miss. A music theorist can judge whether an AI's analysis of counterpoint is actually correct. That domain expertise is exactly what Handshake AI pays for — and pays well.

Related Reading: What Is AI Model Training and Why Do Companies Pay Humans to Do It?

Who Can Join the Handshake AI Fellowship?

Eligibility Requirements

The Handshake AI program is open to U.S.-based bachelor's, master's, and doctoral students, candidates, and graduates with valid work or training authorization. This includes F-1/OPT, J-1, and H-1B visa holders. Participants are responsible for confirming eligibility with their visa sponsor or institution prior to signing up.

A few specifics worth knowing:

F-1 students: CPT and standard OPT are both supported. STEM OPT requiring an I-983 is currently not supported.
Location: You must be physically based in the United States.
Degree levels: Associate's, bachelor's, master's, and PhD holders are all eligible, but higher-tier projects typically require graduate-level credentials.
Non-degree holders: In some cases, individuals without a formal degree but with demonstrated subject-matter expertise may be considered on a project-by-project basis.

Two Participation Tiers

Generalist Fellows are open to associate's and bachelor's degree holders. No specific domain expertise is required. Work involves evaluating and ranking AI-generated content with strong attention to detail and the ability to follow detailed rubrics.

Domain Specialist Fellows (MOVE — Model Validation Experts) are targeted at master's and PhD holders. These are remote, part-time positions focused on evaluating and improving AI models within specific academic disciplines, and they come with significantly higher pay rates.

Related Reading: Best Remote Part-Time Jobs for Graduate Students in 2026

What Tasks Will You Actually Do?

Tasks vary by project and your level of expertise. In general, they fall into four main categories.

Prompt Design involves writing high-quality, domain-specific prompts that challenge an AI model's reasoning. A mathematics fellow, for example, might craft multi-step proof problems that the model has to solve and explain in full.

Response Evaluation involves reviewing AI-generated answers and rating them for accuracy, coherence, and depth. This is not subjective work — you follow detailed rubrics and evaluation guidelines provided for each project.

Comparative Ranking means being presented with two or more model outputs and selecting which is better, with a written explanation based on specific evaluation criteria.

Multimedia Assessment is available on some generalist projects and involves evaluating image, video, or audio content generated or processed by AI systems.

One rule applies across all tasks without exception: the use of external AI tools is strictly prohibited. You cannot use ChatGPT, Claude, or any other AI assistant for prompt ideation, response writing, or editing. Your human judgment and expertise are the entire point of the exercise. Using AI to assist with the work defeats its purpose and can result in removal from the program.

Related Reading: AI Annotation Jobs Explained: What You Actually Do and How Much You Earn

How Much Does the Handshake AI Fellowship Pay?

This is where the program genuinely stands out from most online work opportunities.

Compensation is either hourly or per-task depending on the project. Rates depend on your education level, the complexity of the domain, and the specific project requirements. You will always see the exact hourly rate and any bonus details before you commit to a project — and those rates never change mid-project.

General pay benchmarks by tier:

Generalist (bachelor's level): approximately $30–$50 per hour
Domain Specialist (master's level): approximately $50–$75 per hour
Senior Specialist / Domain Lead (PhD): up to $100 per hour or more on select projects

What Time Is Actually Compensated?

For hourly projects, fellows are paid for time spent preparing tasks, completing tasks, and revising or resubmitting tasks when revisions are requested.

The following are not compensable: optional office hours, community platform activity on Slack or Pronto, administrative email correspondence, and time lost to technical issues outside your control. Always confirm with your project lead before logging time for any activity you're unsure about.

Is Onboarding Paid?

Yes, in most cases. Each project determines its own onboarding requirements, and fellows are compensated for project-specific onboarding time. Upon acceptance, every fellow completes a general onboarding flow consisting of approximately two hours of learning modules, a knowledge quiz, and a live practice task with reviewer feedback. This time is compensated.

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

How to Apply: Step by Step

Step 1 — Create Your Profile

Go to the Handshake AI Fellowship page and submit your profile. You'll enter your academic background, area of expertise, and availability. This profile is how Handshake matches you to relevant projects, so be as specific as possible about your subfield and research focus.

Step 2 — Complete the Assessment

All applicants must complete an initial assessment. Not everyone will advance past this stage. The assessment tests your domain knowledge and your ability to follow detailed instructions — both of which are critical to the actual work. Take it seriously.

Step 3 — Complete the Onboarding Checklist

If your assessment passes review, you'll be invited to continue. This stage involves completing a tax form, providing your Social Security Number, uploading work authorization documents, and verifying your identity through the payment platform. Completing this checklist in full allows Handshake to process your payments quickly once a project begins.

Step 4 — Wait for a Project Match

Projects are launched based on partner needs, so timing varies. If your background matches an active project, you'll receive an invitation via email. Keep a close watch on your inbox, including your spam and junk folders — project invitations are time-sensitive and come by email rather than through the platform dashboard.

Step 5 — Accept, Onboard, and Start Working

Once matched, you'll receive project-specific onboarding materials. After completing any required training, you can begin contributing tasks, logging eligible time, and submitting work on your own schedule.

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

Payment Methods, Payout Schedule, and Tax Obligations

How Payments Are Processed

Depending on your project assignment and when you joined the platform, payments are processed through either Deel or Stripe.

Deel is the primary payment and contract platform used across most projects. During setup, you'll provide your legal name, date of birth, and Social Security Number. This information is private and used solely for payment and tax compliance purposes. Setup takes approximately five minutes.

Stripe is used on some newer projects. After completing your first task, you'll receive a prompt to set up your payout information through Stripe. You can also set up or update your payment method at any time in the Payments tab of your Handshake AI dashboard.

When Do Payments Arrive?

Handshake sends payouts to partner pay platforms by Wednesday afternoon each week. Each payout covers the work period from Monday at 12:00 AM PST to Sunday at 11:59 PM PST of the previous week. It typically takes up to 48 hours for the payment to process and become available for withdrawal.

In practical terms: work completed this week is paid out by the following Wednesday, with funds generally available for withdrawal by Friday. If expected payments have not arrived by noon PST on Friday, you can submit a payment dispute through the Handshake AI Help Center.

Taxes and Independent Contractor Status

Fellows are engaged as independent contractors. This means no taxes are withheld from your payments. You are responsible for reporting your earnings, filing your own taxes, and making quarterly estimated tax payments if your income is consistent. Handshake generates a Form 1099 automatically through Deel for any fellow who earns above the IRS reporting threshold within a calendar year.

A practical rule of thumb: set aside 25–30% of your earnings for federal and state self-employment tax obligations to avoid surprises at tax time.

Related Reading: How to File Taxes as a 1099 Independent Contractor: A Simple Guide for Freelancers

Related Reading: Deel vs Stripe for Freelancers: Which Payment Platform Pays Faster?

Time Commitment and Schedule Flexibility

All Handshake AI projects are part-time and asynchronous. There is no required minimum number of hours per week. You decide how much time you want to contribute, though most projects cap weekly hours at 40 to maintain task quality and consistency. Many fellows contribute between 10 and 20 hours per week.

Because work is asynchronous, you can log in and complete tasks whenever your schedule allows. There are no fixed shifts or required online windows. This makes the program genuinely compatible with graduate coursework, full-time employment, or other part-time commitments.

The Referral Program

Active fellows in good standing can refer new candidates to the program and earn referral bonuses. Bonuses are paid monthly through your existing payment method on the Handshake AI platform. The exact bonus amount for your current project is listed in your Handshake AI dashboard under the Referrals tab.

Is the Handshake AI Fellowship Legitimate?

Yes. Handshake was founded in 2014 and has grown into the leading early-talent hiring platform in the United States, serving over 18 million students and partnering with thousands of employers ranging from Fortune 500 companies to federal government agencies. The AI Fellowship is a fully supported program within the same organization, backed by the same infrastructure and compliance standards.

The program works with confidential AI lab partners — names that are not publicly disclosed due to contractual agreements — described as leading frontier labs focused on general-purpose LLM development. Projects are internally reviewed before launch, tasks are monitored for clarity and fairness, and contributors come from a wide range of institutions and disciplines.

Tips for Getting Selected and Staying Active

Be specific on your profile. List your exact degree, subfield, and any research concentrations. Project matching is done based on expertise alignment — vague profiles get passed over.

Take the assessment carefully. The most common reason applications stall is misreading or rushing the assessment. Read every instruction completely before answering. Missing a qualifier or misjudging a rubric will cost you the placement.

Monitor your email consistently. Project invitations arrive without advance notice. A missed invite can mean losing access to that project cycle entirely.

Stay active in the community platform. Pronto is the primary communication hub for active fellows. Project leads post updates, clarifications, and task availability notices there. Staying engaged keeps you informed and visible.

Log only eligible time. Logging non-compensable activities — optional meetings, platform community interactions — can trigger payment disputes or affect your standing in the program. When in doubt, ask your project lead before logging.

Who This Program Is Not For

The Handshake AI Fellowship is not a fit for everyone. It may not work for you if you are based outside the United States, are on STEM OPT requiring an I-983, need a guaranteed predictable income stream, want to use AI tools to assist with tasks, or are seeking formal W-2 employment rather than independent contractor work.

For those who do qualify — particularly graduate students, recent doctoral graduates, and professionals with deep domain expertise — few remote part-time opportunities offer this combination of hourly rates, schedule flexibility, and meaningful work.

Frequently Asked Questions

Do I need prior AI experience to join? No. Handshake explicitly states that no AI experience is required. What matters is domain expertise, attention to detail, and the ability to follow detailed instructions. All necessary training is provided through the onboarding process.

How much does the Handshake AI Fellowship pay per hour? Rates range from approximately $30 per hour for generalist roles to over $100 per hour for senior PhD-level domain specialist positions. Rates are published upfront before you accept any project and do not change mid-project.

How and when do I get paid by Handshake AI? Payouts are processed weekly via Deel or Stripe. Work from the previous Monday through Sunday is paid out by the following Wednesday, with funds typically available for withdrawal by Friday.

Is the Handshake AI Fellowship open to international students in 2026? It is open to U.S.-based participants with valid U.S. work authorization. F-1 students on CPT or standard OPT are eligible. STEM OPT students with an I-983 requirement are not currently supported.

How long does it take to get matched with a Handshake AI project? There is no set timeline. Some fellows are matched within weeks; others wait several months depending on whether an active project aligns with their expertise. Completing your onboarding checklist fully and keeping your profile current gives you the best chance of a faster match.

Can I work on multiple Handshake AI projects at the same time? It depends on the project. Whether you can stack concurrent assignments varies by project terms. Your assigned project lead can clarify this before you commit.

Are Handshake AI Fellowship earnings taxable? Yes. Fellows are paid as independent contractors. Taxes are not withheld. You are responsible for reporting income and paying estimated quarterly taxes as applicable.

What should I do if I have a Handshake AI payment dispute? If your expected payment has not arrived by noon PST on Friday, submit a payment dispute form through the Handshake AI Help Center. The support team resolves disputes through Deel or Stripe depending on your assigned platform.

Can I refer others to the Handshake AI Fellowship and earn a bonus? Yes. Active fellows in good standing can refer new applicants and earn monthly referral bonuses paid through their existing payment method. Bonus amounts are listed in the Referrals tab of your platform dashboard.

Which AI labs does Handshake partner with? Partner names are kept confidential under contractual agreements. They are described as leading frontier AI labs focused on developing general-purpose large language models.

Disclosure: This article is independently researched and is not sponsored by or affiliated with Handshake. Eligibility criteria, pay rates, and program details are subject to change. Always verify current information directly through Handshake's official Help Center before applying.

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

What Is the Handshake AI Fellowship?