How to apply and start your first AI training project

Last updated: October 21, 2026

Applying takes 1 to 2 hours. You fill in a profile, upload a CV and usually do an AI interview or a test. If your profile fits, you are placed on one or more projects. That can take a few days to a few weeks, because it depends on which projects are open.

What should I get ready first?

Step 1: Your profile and CV

Platforms match you to projects based on what you write, so be specific:

• Write "6 years of audit at a Big Four firm, IFRS 15 and IFRS 16, banking clients" instead of "finance professional".
• Name your specialties, such as LBO modeling, IFRS 9, transfer pricing, SaaS FP&A or credit research.
• Write out qualifications in full, for example "ACCA (Association of Chartered Certified Accountants)". Local titles like "RA" or "WP" need the full name too.
• Add tools: Excel, Bloomberg, FactSet, Capital IQ, SAP, and Python if you use it.
• Mention languages. Some finance projects need German, French, Spanish, Japanese or other languages.

Step 2: The AI interview or test

Most platforms start with an interview led by an AI, by video or voice. It is a structured interview about your own experience. A few platforms use a written test instead.

Some links on this page are referral links. Disclosure

Finance applicants report questions like "Describe a time when you had to work on a project with financial analysis", followed by questions on your answer, for example on enterprise value, working capital or how a deal works day to day.

Tips:

• Do a practice interview first. Mercor and micro1 both offer one for free.
• Answer with real examples from your work, such as a deal, a model, an audit issue or a judgment call. The AI asks follow-up questions on what you say, and a concrete example shows your level better.
• Describe the situation, what you did and the result, in about 45 to 90 seconds per answer.
• Pause when you're done. The AI takes silence as the end of your answer.
• Answer without AI help. Platforms pay for your own judgment, and training AI with AI defeats the purpose of the work. In our experience, it is the most common reason experts are removed from projects or platforms. Some projects allow no AI tools at all, not even for grammar.

After the interview, some roles add a paid work trial. Treat every test and trial like client work. Show your reasoning, mention alternatives you rejected and flag the risks or exceptions an expert would check.

Step 3: ID check, payment and contract

On many platforms you verify your identity and set up payment while you apply, before you know if you're selected. The contract follows when you're placed on a project.

• You verify your identity, often through a provider such as Persona or Deel.
• You set up a payment account with Stripe, Deel, PayPal or Wise, depending on the platform and your country. Some platforms ask for this only after your first task.
• You sign a contractor agreement with confidentiality terms. Read it before you sign.
• Some platforms ask non-US contractors for a W-8BEN, a US tax form confirming you're not a US taxpayer. You still declare the income in your own country.

Step 4: Waiting for a project

A decision after the interview usually takes a few hours to two weeks. A first project can take a few days to a few weeks, or even months.

Silence usually means no project needs your profile right now. Projects open and close based on what AI companies need, and each one looks for a specific mix of expertise, country and available hours. Keep your profile up to date and apply to roles that match your specialism, on more than one platform.

Step 5: Your first project

Each project works differently, with its own rate, tasks, area of finance and hours. What usually happens:

• You get access to the project tools, often Slack and a task platform.
• You read the guidelines, often 10 to 30 pages. Read them closely. This is an important step to succeed in the project.
• You may need to do a few training or qualification tasks.
• You start real tasks. A reviewer checks your work and gives feedback.

Tips for a good start:

• Follow the guidelines exactly, even where you'd work differently at your job.
• Put quality before speed. Low scores can end your place on a project.
• Track only the time you work. On hourly projects the time tracker may take screenshots, and idle time can be deducted.
• Ask questions in the project channel.
• Never paste task content into ChatGPT or another AI tool, and never use employer or client data.

Frequently asked questions