Software Engineer - AI Code Evaluation

About the role

Engineers assess and train next generation AI models by judging generated code and designing challenging programming problems with verified solutions.

What you will do

• Assess AI generated code for correctness, quality, maintainability and sound design
• Write challenging coding problems with verified reference solutions and thorough test cases
• Review model reasoning on architecture, debugging, algorithms and system design
• Find logic errors, inefficient code, edge cases and architectural weaknesses
• Give detailed technical feedback and explanations to improve model performance
• Apply consistent evaluation criteria for accurate and reliable assessments

What they look for

• At least three years of professional experience in data science or software engineering
• At least one year at a leading technology, finance or research company
• Strong programming and software engineering fundamentals
• Ability to judge code correctness, find bugs and assess design decisions
• Strong analytical thinking and clear written English for technical feedback