The work
Create and assess software engineering examples used to train and evaluate AI models. The duties and qualifications focus on broader software engineering evaluation, so review them before applying.
- Curate code examples and write precise solutions for training datasets.
- Correct and refine code in Python, C/C++, Rust, Go, Java, and JavaScript, including ReactJS.
- Evaluate AI-generated code for correctness, performance, scalability, and reliability, especially in systems, infrastructure, and backend work.
- Build verification agents to identify error patterns and assess infrastructure code.
- Evaluate model capabilities in prototyping, architecture, API design, production implementation, monitoring, and maintenance.
- Design ways to automatically check solutions to software engineering tasks.
What you need and the schedule
- At least three years of software engineering experience.
- Strong systems programming, infrastructure, or backend development skills, especially with Python, C/C++, Rust, or Go.
- Experience building and deploying scalable, production-grade software.
- A strong grasp of software architecture, development, debugging, and code-quality assessment, plus the ability to explain technical evaluations clearly.
- Experience with advanced AI systems or organizations developing intelligent systems is helpful. Engineering skill and performance are prioritized alongside formal academic background.
- Fluent in English.
- Open to candidates based in the United States; remote contractor engagement.
- Part-time schedule of 20+ hours per week. The engagement is planned for one month, with possible extensions based on performance and fit.