How to get accepted on Outlier: assessment and CV guide

The onboarding gate is everything

On Outlier, your onboarding assessment determines which projects you can access and at what rate. Treat it as a job interview, not a formality. Most people fail the assessment on their first try, not because they lack ability but because they do not understand what the test is actually measuring. Outlier rewards precision, not speed. It punishes vagueness. And it hates "good enough" answers.

Outlier is operated by Scale AI and connects freelance contributors with AI companies. The work involves evaluating AI responses, writing prompts, ranking answers, editing text, and providing domain-specific feedback. Pay ranges from $15 to $50 per hour depending on your domain and location, with software engineering and specialist fields like medicine and law paying the most.

Apply to a specific role, not the generic sign-up

The biggest mistake new applicants make is using the generic "Apply" button on the Outlier homepage. Instead, find a specific job listing that matches your skills and apply through that. Outlier lists opportunities across software engineering, coding, writing, languages, mathematics, science, and specialist domains. Applying to a specific role means your profile is evaluated against that role's requirements, rather than being dumped into a general pool.

When you create your account, your personal and professional information must be accurate. Outlier uses your profile and resume to evaluate whether you meet the requirements for a particular domain. A vague profile that says "Software Developer" will score lower than one that lists React.js, JavaScript, TypeScript, Node.js, REST APIs, and front-end development as distinct skills.

Choose your domain wisely

Outlier lets you pick a domain for your assessment. This is where most people go wrong. They pick "Generalist" because it sounds easy. That is a trap. Generalist questions are broad and ambiguous, and harder to nail. Pick a domain where you have genuine knowledge. If you have worked in operations, choose the operations track. If you know finance, pick that. The assessment is easier when you can spot errors in your own field.

You can select up to 10 skills, but only choose those where you could perform well today, not skills you plan to learn. Outlier's matching engine compares your self-selected skills against verified signals like LinkedIn, GitHub, and academic email. Claiming skills you cannot demonstrate under timed conditions will result in failed screenings and potential account flags.

Prepare your CV and profile before you start

Outlier requires a current resume highlighting your expertise and a LinkedIn profile showing your education and work experience. Your resume should be current, easy to read, focused on relevant skills, and free from exaggerated claims. For technical roles, highlight your programming languages, frameworks, projects, and professional experience. For writing or language roles, highlight relevant writing, editing, translation, or communication experience.

Update your resume to reflect your strongest domain experience from the past three to five years. Use specific language. "Developed a Python data pipeline for financial reporting" is stronger than "proficient in Python". Connect your LinkedIn profile before starting the assessment, because the system cross-references it automatically and a populated profile increases credibility for technical and academic skill claims. If you are a software engineer, push recent code to GitHub, as Outlier's system validates technical skills through GitHub profile analysis alongside the live coding test.

If you have no direct AI training experience, do not try to fake it. Instead, emphasise transferable skills. Outlier values attention to detail, clear written English, and reliability. Frame your experience in terms of accuracy, throughput, and quality, because these are the metrics that matter in AI training work. If you are a domain expert in law, medicine, finance, or education, highlight your professional credentials and your ability to evaluate information critically.

How to pass the skills assessment

The assessment is the gatekeeper. You get a timed test that matches the domain you selected. If you pick coding, expect LeetCode-style problems. If you pick writing, expect to edit paragraphs and rank responses. The assessment is hard by design. Outlier wants people who can do the work without handholding.

Read the rubric before you answer anything. Outlier provides a scoring rubric for each task. Rate each criterion separately. Do not average them in your head. If a response is factually correct but poorly written, say that. Do not give it a middle score because it balances out.

Be brutally specific in your feedback. Vague feedback gets you rejected. If the AI response says "Python is a programming language used for data science", do not write "This is correct". Write "This statement is factually accurate but incomplete. The response should mention these use cases." Quote the exact part of the response you are evaluating, explain what is wrong, and suggest a fix. Three sentences maximum per issue.

Take notes throughout the onboarding and assessment process. From training modules to task instructions, keeping a written record helps you reference guidelines while working and reduces errors. Many failures come from inconsistent labelling or missed edge cases, so brush up on basic reasoning and attention-to-detail skills before you start.

Practical tips for assessment day

Do the assessment in one sitting. Do not pause for hours. Use a quiet room with no distractions. Read each prompt twice, once to understand it and once to find the trap. If you are unsure about a fact, do not guess. Mark it as "cannot verify" in your feedback. Keep your tone neutral. Do not be sarcastic or emotional. Check your spelling, because Outlier flags sloppy writing.

Take your time. The biggest mistake is rushing. Take twice as long as you think you need. Focus on factual accuracy and logical consistency over creativity. If you fail, you can wait 30 days and try again, potentially with a different domain.

After acceptance

Passing the assessment does not mean you start working immediately. You go into a pool. When a project needs people with your skills, you get an invite. This can take days or weeks. Each project has its own onboarding, where you read guidelines, complete practice tasks, and get graded.

Your first few submissions set your quality rating. Invest extra care here, because early ratings compound into access to higher-paying batches. Do not rush for volume in week one. Do 10 to 15 hours, learn the system, and understand what gets approved and what gets rejected. Speed comes later.

Read the project guidelines twice. Most rejected work comes from ignoring the guidelines. Read them before you start working, then read them again after your first few tasks. Check for work regularly, because Outlier does not email you when new tasks appear. If you wait, other workers grab the tasks first.

Every task you submit gets reviewed by senior raters or automated checks. If your quality drops, you can be removed from the project without warning. Specialise rather than generalise. Generalists earn less and get fewer projects. If you can do maths or coding, focus there, because the pay is better and the competition is smaller.