Software Engineer - AI Code Evaluation

About the role

Gramian Consultancy, a boutique consultancy specialising in IT professional services and engineering talent solutions, is looking for Software Engineers to help train and evaluate next-generation AI models on real-world software engineering tasks. You will assess AI-generated code for correctness, quality and design, create challenging programming problems with verified reference solutions, and evaluate model reasoning across architecture, debugging and system design. The role suits technically strong professionals who can give clear, actionable feedback.

Key responsibilities

• Evaluate AI-generated code for functional correctness, code quality, maintainability and sound software design.
• Create challenging coding problems, verified reference solutions and comprehensive test cases.
• Review AI reasoning on software architecture, debugging, algorithms and system design.
• Identify logical errors, inefficient implementations, edge cases and architectural weaknesses.
• Provide detailed technical feedback and explanations to improve AI model performance.
• Apply consistent evaluation criteria to ensure accurate and reliable assessments.

Required profile

• At least 3 years of overall professional experience in Data Science or Software Engineering.
• At least 1 year of professional experience at a leading technology, finance or research company.
• Bachelor's degree in Data Science, Software Engineering, Computer Science or a related field from a university in the United States, Canada, Australia, New Zealand or the United Kingdom.
• Strong written English and the ability to communicate technical feedback clearly.

Required skills

• Strong programming and software engineering fundamentals.
• Ability to evaluate code correctness, identify bugs and assess software design decisions.
• Strong analytical thinking and the ability to solve complex technical problems.