CrowdGen by Appen

What CrowdGen is

CrowdGen is where Appen recruits and manages the people who do its data work. Appen has worked on language and AI data for decades, from early speech recognition to large language models, and is listed on the Australian Securities Exchange under the code APX. Through CrowdGen, contributors work on projects for Appen's customers: evaluating model answers, annotating text, audio and images, writing data and checking translations.

How it works

The site describes four steps: apply and get verified, get matched to projects in your field, contribute, and build a quality record. Its FAQ says that "most of our projects have a qualification process", so you usually cannot start paid tasks on the day you register. Each project has its own guidelines and test, and some ask for specific languages, locations or qualifications.

Who it suits

CrowdGen recruits across a wide range: linguists and native speakers of many languages, people with general reading and writing skills, software engineers, and specialists in fields such as finance, law and medicine. Language projects are often the most open to people outside the US and Europe, because AI companies need data in many languages and accents.

Pay and payment

The site states that pay rates depend on location and skills, and that rates are shown before you begin. Earnings are mostly stated in US dollars and paid in your local currency, with withdrawals through bank transfer, PayPal, Payoneer, Airtm and other methods depending on where you live. The FAQ also says: "We will never ask you to pay anything to join our projects."

Honest advice

• Qualification tests are where most applicants drop out. Read the project guidelines twice before you start the test.
• Projects open and close. Keep your profile complete, including languages and skills, so you are matched when one opens.
• Fake "Appen recruiters" on messaging apps ask for fees. Apply only through the official site.

Best for: multilingual students and graduates, and specialists, who want project-based AI data work from many countries.