Swift Engineer (5+ YOE) – AI / LLM Code Evaluation (Remote, Contract)

Company: Mercor.
Type: Contract (Full-time or Part-time).
Location: Remote (Worldwide).
Language: Professional English required.

Compensation:
• USD $30 – $90/hour (depending on experience & evaluation performance).
• Weekly payments via Stripe or Wise.
• Flexible workload (project-based, scalable hours).

Mission:
• Work directly with leading AI teams to improve how large language models reason about code, systems design, and technical problem-solving.
• You will evaluate and refine AI-generated responses, making them more accurate, reliable, and aligned with real-world engineering standards.

Responsibilities:
• Evaluate AI-generated answers to coding and system design problems.
• Execute and validate code outputs.
• Identify bugs, inefficiencies, and incorrect reasoning.
• Assess code quality & readability.
• Assess algorithmic correctness.
• Assess system design logic.
• Annotate responses with structured, actionable feedback.
• Follow defined evaluation frameworks and quality benchmarks.

Required Skills:Core:
• Swift (expert level).
• Software Engineering (5+ years).
• Data Structures & Algorithms.
• Systems Design.
• Debugging & Code Review.
• Problem Solving (Medium–Hard level).
Technical:
• Code Execution & Testing.
• API Design & Backend Logic.
• Performance Optimization.
• Version Control (Git).
AI / Evaluation Context:
• Experience using LLMs in development workflows.
• Ability to evaluate reasoning, not just outputs.

Nice-to-Have Skills:
• RLHF / AI Model Evaluation.
• Competitive Programming.
• Open-source contributions (merged PRs).
• Multi-language experience (Python, JS, etc.).
• Technical writing / explaining complex concepts.

Ideal Candidate:
• Degree in Computer Science or related field (BS/MS/PhD).
• Strong real-world engineering background.
• Detail-oriented and highly analytical.
• Comfortable identifying subtle logic flaws and edge cases.
• Able to work independently in async environments.

What You Will Achieve:
• Improve the quality and reasoning of AI-generated code.
• Influence how AI systems assist developers globally.
• Deliver high-quality evaluation outputs that directly impact model performance.

Location: Remote - Anywhere

Skills required for this job:
• AI model evaluation
• API design
• Algorithm development
• Code review
• Data science
• Data structures
• Debugging
• Git
• JavaScript
• Large language model (LLM)
• Performance optimization
• Python
• Reinforcement learning from human feedback (RLHF)
• Software engineering
• Swift
• Systems design
• Technical writing
• Testing