The Work
You will create challenging backend coding tasks for AI systems and build deterministic verifiers. The tasks should reflect realistic feature work, debugging, and bug-fix situations, while the verifiers must accept valid solutions and reject solutions that miss the specifications.
The role title identifies backend security engineering, while the duties focus on coding tasks and verifiers. Review the duties and qualifications before applying.
- Create backend tasks involving feature development, debugging, and bug fixes.
- Build verifiers that distinguish valid solutions from invalid ones.
- Assess code quality, performance, testing, refactoring, algorithms, data structures, and scalable services.
- Use secure coding and security review practices when relevant.
What It Pays And Takes
This is a remote contractor role with flexible scheduling.
- Pay: $30 to $100 per hour equivalent, paid per task that meets the project specifications.
- Schedule: Approximately 15 hours per week with flexible scheduling.
- Language: Professional fluency in English.
- Backend experience in Python, Java, Node.js, Go, Rust, TypeScript, C++, or a similar language.
- Experience building scalable services and APIs, debugging software, resolving bottlenecks, implementing complex features, and reviewing code.
- Strong analytical, problem-solving, and technical communication skills.
- Helpful background includes secure coding, application security, penetration testing, vulnerability assessment, or security-focused code review.
- Familiarity with OWASP or CWE guidance and large-scale, performant software is helpful.
How it works
You apply here in about two minutes and finish on the hiring platform. The next step is a short AI interview (technical roles include a coding challenge). Passing it is how you qualify for paid tasks, so take it soon after you are invited.
About AI Training Work
AI training is the human work behind artificial intelligence. People create and review examples, including code and model responses, so AI systems can learn to produce more useful results.
Experienced contributors are paid for applying technical judgment to difficult tasks, such as testing software solutions, checking requirements, and evaluating code quality.