Handshake AI: Jobs, Pay, Requirements, and How It Works in 2026
Handshake AI is a paid, project-based program operated by Handshake that recruits students, graduates, professionals, and subject-matter experts to help train and evaluate artificial intelligence models.
Participants may write prompts, compare AI-generated responses, correct technical content, or review outputs using their academic or professional expertise. Most current Fellowship opportunities are remote and part-time, but applicants must meet project-specific qualifications and Handshake’s location and work-authorization rules.
The advertised rates can be attractive. However, they are generally presented as project-specific maximums, not standard pay or guaranteed income. Applying also does not guarantee that you will be matched with a project.
Handshake AI at a glance
- Program: Paid AI training and evaluation projects
- Operator: Stryder Corp., doing business as Handshake
- Work model: Primarily part-time contract projects
- Location: The main Fellowship currently requires participants to be US-based
- Remote work: Available for many Fellowship projects, but not every Handshake opportunity
- Compensation: Hourly or per-task, depending on the project
- Availability: No guaranteed project match or fixed invitation timeline
- Best suited to: Eligible applicants comfortable with variable, project-based work
- Main caution: Do not treat an advertised maximum rate as expected or guaranteed earnings
Information checked: September 12, 2026. Opportunities, rates, eligibility requirements, and project availability can change. Verify the current listing and Project Terms before applying.
What Is Handshake AI?
Handshake AI is an initiative within the wider Handshake career platform. Handshake’s privacy policy identifies the company responsible for the service as Stryder Corp., doing business as Handshake.
The company describes the Handshake AI Fellowship as a program through which selected applicants complete paid projects that help train and evaluate AI systems. Handshake connects participants, known as Fellows, with work created for AI labs and other participating organizations.
The word “Fellowship” can make the program sound like one standardized academic placement. In practice, it encompasses different contract projects for generalists, students, researchers, experienced professionals, and highly specialized experts.
It is also important to distinguish among three related services:
- The main Handshake platform lists internships and jobs from outside employers.
- The Handshake AI Fellowship offers paid AI training and evaluation projects.
- Handshake also uses the Handshake AI name for an AI interview service.
This article focuses on the paid AI-work and Fellowship ecosystem, not ordinary jobs that outside employers post on Handshake.
What Do Handshake AI Fellows Actually Do?
The work depends on the project. Handshake’s official Fellowship overview identifies activities such as:
- Creating prompts within a Fellow’s field
- Evaluating or comparing AI-generated responses
- Correcting and improving content
- Supporting corrections with academic citations
Current specialist listings provide more concrete examples. Handshake’s software-engineering opportunity asks participants to design job-related coding questions and review AI-generated answers for correctness, efficiency, clarity, and alignment with real-world engineering practices.
A generalist project may instead focus on reasoning, attention to detail, testing large language models, and following detailed instructions. Applicants should therefore evaluate the underlying responsibilities, not rely solely on a listing’s title.
Human reviewers are useful because a model can produce a fluent answer that contains factual errors, faulty reasoning, weak evidence, or unrealistic professional advice. A qualified reviewer can identify those problems and provide more reliable feedback.
Most Fellows are not building machine-learning systems. They are producing or evaluating the human feedback, examples, and expert material used to test or improve those systems.
How Does Handshake AI Work?
The process involves more than submitting an application and immediately receiving tasks.
Handshake’s current documentation describes a general path from account creation to project matching:
- Create or access a Handshake account.
- Complete your AI-work profile.
- Confirm work-authorization information.
- Submit the required application details.
- Wait for a matching project.
- Express interest if invited.
- Complete any required assessment or screening.
- Finish project-specific onboarding and training.
- Receive approval to start working.
- Complete eligible tasks under the Project Terms.
- Receive payment through the configured payout system.
The details can differ by project. Being eligible, applying, passing an assessment, and receiving paid work are separate stages.
Applying and Creating a Profile
Applicants can start from the Fellowship page or Handshake’s current AI opportunities page.
According to Handshake’s application instructions, updated July 31, 2026, the process includes:
- Signing in or creating a Handshake account
- Connecting a school or joining the open network
- Verifying a phone number
- Answering US work-authorization questions
- Describing your education, experience, and expertise
- Uploading a résumé in PDF format
- Accepting the applicable terms, privacy policy, and contractor agreement
Once the profile is complete, the applicant waits for a potential match. Handshake does not publish a guaranteed response or placement timeline.
Assessments and Qualification
There is no single assessment for every Handshake AI job.
The generalist listing says applicants receive brief training and complete one assessment before starting. The software-engineering listing requires a coding assessment. Other opportunities may request credentials, portfolios, professional samples, additional screening, or project-specific tests.
An applicant who passes one assessment should not assume that they are automatically qualified for every future project. Each project can have its own selection and onboarding requirements.
Project Matching and Availability
Project availability is one of the most important limitations.
Handshake’s June 2026 application-to-project guide says projects launch according to the needs of participating AI labs. There is no fixed timeline for receiving an invitation.
If an applicant’s profile matches an available project, Handshake may send an interest email containing:
- A project description
- Estimated project length
- Expected time commitment
- Compensation information
- Participation requirements
- Requests for additional materials
Expressing interest does not confirm selection. Candidates are reviewed again, and those chosen receive a separate acceptance message.
Qualification therefore does not equal continuous access to work. A person may complete a project and then wait for another suitable assignment.
Project Onboarding and Training
Selected Fellows must complete the requirements shown in the project workspace before beginning paid tasks.
Training may cover:
- Project responsibilities
- Deliverables and quality standards
- Task instructions
- Permitted tools
- The work environment
- Time tracking
- Review and revision procedures
Handshake’s educational-partner guidance says onboarding requirements vary by project. Applicants should confirm which onboarding activities are compensated under their current Project Terms rather than assuming that all preparation is payable.
Completing Tasks and Getting Paid
Handshake says most projects use either hourly or per-task compensation.
Its earnings guidance says hourly projects may compensate Fellows for preparing, submitting, revising, and resubmitting tasks. However, some tasks have a maximum handling time.
When a time cap applies, payable time is limited to that cap even if the task takes longer. This means the headline hourly rate alone may not reveal the effective return for a difficult assignment.
Fellows must also complete tax and payout setup. Handshake’s payment-processing guide identifies Stripe as its payout provider.
The guide says Handshake typically sends payouts to its payment partners by Wednesday afternoon for the previous Monday-to-Sunday period. Processing may take approximately 48 hours. Some projects use a different schedule, which should be stated in their Project Terms.
Handshake also provides a payment-dispute form for Fellows whose expected funds have not arrived after the published processing window.
What Handshake AI Jobs Are Available?
Handshake AI opportunities change as projects open, reach capacity, or end. The public pages displayed both generalist and specialist categories when checked on September 12, 2026.
Examples included:
- Generalist
- AI Evaluation Specialist
- Energy Professional
- FP&A Analyst
- Game Developer/Designer
- PCB Tool Specialist
- Philosophy Expert
- Software Engineer
- Claims Specialist
These examples should not be treated as permanently available positions.
The job titles can also resemble conventional occupations even when the tasks are different. The software-engineering opportunity, for instance, involves evaluating AI-generated code and technical content. Handshake explicitly states that it is not a traditional full-time software-engineering role.
A claims-related listing similarly asks experienced claims professionals to review AI-generated material involving claims processes, policy language, and customer communication. It is AI evaluation informed by insurance expertise, not an ordinary claims-adjusting position.
Before applying, examine:
- The actual tasks
- Mandatory qualifications
- The compensation model
- Any assessment requirements
- Location and work-authorization restrictions
- Whether placement depends on future demand
How Much Does Handshake AI Pay?
There is no universal Handshake AI hourly rate or salary.
Official pages advertise different maximum rates for different opportunities. The following examples were visible on September 12, 2026:
| Opportunity | Officially advertised compensation | Listed education |
| Generalist | Up to $30 per hour, depending on project | Associate’s or bachelor’s on the listing |
| AI Evaluation Specialist | Up to $40 per hour | Associate’s or bachelor’s |
| Software Engineer | Up to $65 per hour, depending on project | Bachelor’s, master’s, or PhD |
| Energy Professional | Up to $80 per hour | Multiple degree levels |
| FP&A Analyst | Up to $80 per hour | Multiple degree levels |
| Philosophy Expert | Up to $120 per hour | PhD or postdoctoral |
| Game Developer/Designer | Up to $125 per hour | Multiple degree levels |
| PCB Tool Specialist | Up to $125 per hour | Multiple degree levels |
These are advertised maximums, not verified average earnings.
“Up to $65 per hour” does not mean every accepted software engineer receives $65 per hour. It also does not show how many tasks will be available, how long the project will last, or whether every task is payable at the maximum rate.
An official Handshake help page for educational partners states that Fellowship rates can vary according to education, expertise, and project. Live listings remain the better source for evaluating an individual opportunity because rates and eligibility conditions can change.
Some projects use per-task rates instead of hourly pay. A per-task offer cannot be meaningfully compared with an hourly listing without knowing:
- How long an acceptable task normally takes
- Whether revisions are compensated
- Whether submissions can be rejected
- Whether a handling-time limit applies
- How much work will actually be available
For most Fellowship opportunities, “Handshake AI salary” is not an accurate description. Handshake describes Fellows as contractors completing project-based work rather than salaried employees.
What Determines Handshake AI Pay?
Official documentation identifies several relevant factors:
- The project
- Subject-matter expertise
- Education level where specified
- Hourly versus per-task compensation
- Task-level time limits
- Quality and acceptance conditions
- The applicable Project Terms
Applicants should save the terms provided when accepting a project. Those terms should be checked for rules governing time tracking, revisions, incentives, rejected work, onboarding, prohibited tools, and payment disputes.
Does Handshake AI Pay $100 an Hour?
Some specialist opportunities advertise maximum rates above $100 per hour. That does not mean every participant earns $100 per hour.
To receive a particular rate, an applicant would need to:
- Meet the listing’s qualifications
- Be matched with an active project
- Pass the selection process
- Accept the offered compensation
- Receive eligible tasks
- Complete the work satisfactorily
- Comply with the Project Terms
Even then, total earnings depend on work volume and project duration.
Can You Realistically Make Money Through Handshake AI?
Selected Fellows can receive payment for eligible work under their project’s compensation and quality conditions. Handshake publishes an established payout process, and worker reviews include reports of paid projects.
What cannot be predicted from the advertised rate is how much any applicant will earn.
There may be no immediate project match, and a completed project does not guarantee another assignment. Handshake AI is therefore more realistically treated as potential supplemental contract income unless and until a Fellow has active, consistent work.
What Are the Handshake AI Requirements?
Handshake AI requirements vary by project. There is no universal qualification profile that applies to every opportunity.
Education
The generalist listing accepts current students or graduates at specified degree levels. Specialist listings may require bachelor’s, master’s, doctoral, or postdoctoral qualifications.
Handshake’s educational-partner guidance says many expert roles require graduate-level knowledge. It also says applicants without formal degrees may sometimes be considered when they can demonstrate relevant subject-matter expertise.
The individual listing determines whether a degree is required.
Professional Experience
Some opportunities require extensive experience.
The software-engineering listing available in September 2026 required at least four years of professional software-engineering experience, excluding internships. It also asked for strong coding skills, production-level development experience, written communication skills, and the ability to follow technical guidelines independently.
The generalist listing did not require specialist professional experience. It emphasized attention to detail, reasoning, comfort with asynchronous work, and the ability to follow instructions.
Technical and Identity Requirements
Handshake does not publish one universal hardware specification for all projects. Confirmed account and onboarding requirements may include:
- Internet access and a supported browser
- Phone-number verification
- A PDF résumé
- Identity verification
- Tax information
- Stripe payout setup
- Access to the project’s work environment
Some projects may introduce additional software, monitoring, time-tracking, or security requirements during onboarding.
Location and Work Authorization
As of September 12, 2026, Handshake’s main Fellowship page states that participants must be based in the United States and have work authorization that Handshake can support.
Handshake says F-1 students eligible for Curricular Practical Training or Optional Practical Training may qualify. STEM OPT is not supported. Students are advised to work with their Designated School Official to determine eligibility.
Some help-center material refers more broadly to international graduate students where visa conditions permit. In context, this appears to concern eligible international students based in the United States rather than unrestricted participation from other countries. Applicants should rely on the latest official program page and obtain individual guidance where necessary.
Remote work does not mean worldwide availability.
Handshake’s privacy policy also states that contractors must be at least 18.
Other Project-Specific Requirements
Depending on the opportunity, an applicant may be asked to provide or complete:
- Educational credentials
- Professional history
- A coding or knowledge assessment
- A portfolio or work sample
- Identity and liveness verification
- A background check
- Project-specific training
- An AI interview
Handshake says AI interviews are normally optional but may be required for certain projects. Its privacy policy states that the company stores and processes the content and recordings of those interviews.
How to Apply for Handshake AI
The current official process is:
- Visit the Handshake AI Fellowship or opportunities page.
- Select a suitable opportunity or use the general sign-up option.
- Sign in or create a Handshake account.
- Connect your school or use the open network where offered.
- Verify your phone number.
- Answer the work-authorization questions accurately.
- Complete your expertise and profile information.
- Upload the requested PDF résumé.
- Read and accept the applicable agreements.
- Wait for a relevant project match.
- Review the invitation, rate, duration, and requirements.
- Complete any assessment or additional screening.
- Finish identity verification, onboarding, and payout setup.
- Begin work only after receiving project approval.
Use accurate credentials and location information. Misrepresenting qualifications, identity, or geographic eligibility could result in rejection, account action, or a payment dispute.
Submitting an application only makes the profile eligible for matching. It does not guarantee acceptance or paid work.
Is Handshake AI Legit?
Handshake AI is a verifiable program operated by an established corporate entity. It has:
- An official program website
- Public opportunity listings
- Application documentation
- A help center
- Payment instructions
- A privacy policy
- Terms of service
- A contractor agreement
- A published payment-dispute procedure
These are strong signs that Handshake AI is an operating AI-work platform rather than an anonymous job offer.
Legitimacy does not mean risk-free participation, guaranteed work, or universally positive contractor experiences.
In March 2026, Business Insider reported that five contractors said their accounts had been suspended and that four were denied compensation after Handshake accused them of contract violations. The contractors disputed those allegations. Handshake declined to comment for the report.
Business Insider also described two legal disputes involving claimed unpaid compensation. One was dismissed by the claimant. A separate case ended with a court awarding a contractor $6,475.
These reports do not establish that every Fellow will experience a payment problem. They do demonstrate why applicants should understand:
- Quality and acceptance rules
- Location restrictions
- Identity requirements
- Permitted and prohibited tools
- Time-tracking expectations
- Account-suspension provisions
- Payment-dispute procedures
The evidence supports a nuanced conclusion: Handshake AI is a legitimate, functioning program, but it has the availability, enforcement, and payment risks associated with project-based platform work.
How Consistent Is Handshake AI Work?
Handshake does not promise consistent projects or a fixed number of hours.
Its official application guide says projects launch according to partner demand and that there is no fixed timeline for an invitation. A qualified Fellow may receive a suitable project, complete it successfully, and then wait for another match.
Reviews on Indeed’s Handshake AI page include individual reports of interesting, well-paid work alongside short projects or limited weekly hours. These are self-reported experiences and cannot establish a platform-wide average.
However, the reported variability is consistent with Handshake’s own description of demand-driven project matching.
Anyone who requires predictable weekly income should not budget around Handshake AI until they have an active project with confirmed work.
Is Handshake AI Remote?
Most current Fellowship opportunities are advertised as remote, part-time, and asynchronous.
For example, the generalist and software-engineering listings allow independent remote work and state that there is no minimum commitment to hours. Handshake promotes scheduling flexibility across its Fellowship program.
Three limitations apply:
- Not every opportunity connected with Handshake AI is necessarily remote. Handshake has also advertised in-person research participation.
- Remote projects may use time tracking or session monitoring.
- The main Fellowship currently requires US residence and supported work authorization.
Always read the location and eligibility fields of the exact opportunity. Do not rely on a social-media post, recruiter message, or another applicant’s circumstances.
Privacy, Monitoring, and Confidentiality
Handshake AI participation can require more personal and professional information than an ordinary job-board profile.
The Handshake privacy policy, updated July 3, 2026, states that contractor data may include:
- Name and contact details
- Date of birth
- Education and professional expertise
- Location and languages
- Government-issued identification
- A selfie and liveness check
- Tax and payment information
- Citizenship and work-authorization data
- Background-check information where applicable
- AI-interview content and recordings
Handshake also says some active projects may monitor work sessions. Monitoring can include periodic screenshots, application activity, keyboard-activity levels, and time spent on tasks.
The policy states that contractors will be informed during project onboarding when monitoring is used and that monitoring is limited to active Handshake-related work sessions. Applicants should still understand what software is installed, what it captures, when it operates, and whether the project can be completed on a separate device or user profile.
The privacy policy permits Handshake to share certain contractor account information with authorized agents of AI labs. Depending on the situation, that may include AI-interview content, transcripts, and AI-generated analysis.
Handshake says it retains personal data as needed to provide its services, support legitimate business purposes, meet legal obligations, maintain security, prevent fraud, resolve disputes, and enforce agreements. This is not a promise that all data will be deleted immediately when a project ends.
Professional Documents Require Additional Caution
Applicants may be asked for portfolios, work samples, or professional material.
In August 2026, Business Insider reported on a Handshake AI document-contribution opportunity offering payment for accepted professional documents. The reported listing required applicants to own the material and be authorized to share it.
Creating a document at work does not necessarily mean you own or may distribute it. Rights may belong to an employer, client, publisher, university, or contracting party.
Documents can also contain:
- Personal information
- Confidential business data
- Trade secrets
- Client material
- Privileged communications
- Internal financial information
- Copyrighted content
- Information covered by a nondisclosure agreement
Do not submit confidential or proprietary material unless you are authorized to do so and the relevant terms clearly allow it.
If ownership or authorization is uncertain, do not upload the document merely because compensation is offered. Seek appropriate professional advice when necessary.
Handshake AI Pros and Cons
| Pros | Cons |
| Official opportunities cover generalist and specialist subjects | Applying does not guarantee a project |
| Many Fellowship projects are remote and flexible | The main program is currently restricted to eligible US-based participants |
| Live listings disclose advertised maximum rates | Maximum rates do not establish typical earnings |
| Some opportunities require no prior AI-training experience | Specialist roles may require advanced education or substantial experience |
| Both hourly and per-task projects may be offered | Time caps and acceptance rules can affect effective earnings |
| Payment and dispute guidance is publicly available | Contractors have reported account and payment disputes |
| Work can provide practical AI-evaluation experience | Project volume and duration may vary |
Is Handshake AI Worth It in 2026?
Handshake AI may be worth applying to if you meet the current eligibility rules, match a relevant opportunity, and understand that the work is project-based.
It is most suitable for applicants who:
- Have relevant academic or professional knowledge
- Can work independently
- Are comfortable with assessments and detailed quality rules
- Want flexible AI-evaluation experience
- Can tolerate gaps between assignments
- Do not need guaranteed weekly hours
- Can protect employer and client information
It may not be suitable for people who:
- Need predictable monthly income
- Are outside the permitted location
- Do not meet current project qualifications
- Prefer long-term employment
- Are uncomfortable with identity verification or monitored work sessions
- Cannot verify their right to share requested documents
The decision should be based on a real project invitation, not the highest rate advertised on the opportunities page.
Before You Apply: A Practical Checklist
Before accepting a Handshake AI project:
- Confirm that the opportunity appears on an official Handshake domain.
- Check the location and work-authorization requirements.
- Verify that you meet every mandatory qualification.
- Save a copy of the listing and advertised rate.
- Read the Project Terms before accepting.
- Determine whether compensation is hourly or per task.
- Check for maximum handling times.
- Confirm whether assessments and onboarding are paid.
- Review prohibited tools and AI-use restrictions.
- Check whether work sessions will be monitored.
- Complete identity and payout steps only through official systems.
- Do not submit confidential employer or client material.
- Do not assume another project will follow the current one.
- Do not budget around projected income until work is assigned.
Final Verdict
Handshake AI is a legitimate, project-based program that pays selected generalists and subject-matter experts to help evaluate and improve AI systems.
Its strongest attractions are remote flexibility, varied areas of expertise, and advertised rates that can be comparatively high for specialized work. Those rates must be interpreted carefully. They are opportunity-specific maximums, not standard Handshake AI pay or proof of predictable earnings.
Project availability is the main practical constraint. Applying, qualifying, or completing one assignment does not guarantee continuing work.
Handshake AI is best suited to eligible US-based applicants who have useful academic or professional knowledge, can work independently, and view the program as potential supplemental contract work. Before joining, verify the current listing, compensation model, work-authorization conditions, Project Terms, monitoring disclosures, and data-handling requirements.
Do not rely on the headline rate until you have received a real project offer with clear payment and task conditions.
