This description is a summary of our understanding of the job description. Click on 'Apply' button to find out more.
Apply Now
Role Description
This role involves creating terminal bench–style problems and tasks involving agents within a competitive programming environment, as part of a hiring effort for its AI lab partner building advanced agent-based infrastructure.
- Develop interactive, terminal-based challenges that test the problem-solving capabilities of autonomous agents in real-time coding scenarios.
Qualifications
- Track record of success in competitive programming contests (e.g., ICPC, Codeforces, LeetCode, AtCoder, HackerRank)
- Strong skills in algorithms, data structures, and systems-level thinking
- Proficient in Python, C++, or Go, and comfortable working in terminal-based coding environments
- Ability to design programming problems that test time complexity, memory constraints, logic correctness, and multi-step reasoning
- Experience with online judge systems, sandboxed environments, or contest infrastructure
- Curiosity about AI agents and interest in how agents can solve real-world programming tasks
- Enjoy creating reproducible, progressive challenges and can abstract real-world systems problems into code-based puzzles
Requirements
- Create terminal bench–style problems and tasks where autonomous agents are challenged to solve competitive programming problems in a live, terminal-driven environment
- Test agent performance across algorithmic domains — such as graph theory, dynamic programming, greedy techniques, and real-time debugging
What You’ll Do
- Design and implement terminal-based coding challenges for agents to solve autonomously
- Build a sandboxed, judge-like environment that runs test cases, checks correctness, and monitors time/memory usage
- Simulate real-world competitive conditions: limited time, incremental hints, multiple submissions, etc.
- Collaborate with AI engineers to define reward functions, failure conditions, and difficulty tiers
- Develop tooling for benchmark resets, output validation, and scenario generation
- Evolve the difficulty of problems from beginner-level syntax and logic to advanced algorithmic pipelines and systems simulations
Why This Role Is Exciting
- Define a new generation of programming benchmarks built specifically for AI agents
- Blend algorithmic design, problem solving, and systems-level realism in one role
- Work on uncharted challenges that push the boundaries of what agents can understand and execute
Pay & Work Structure
- Classified as an hourly contractor to Mercor
- Paid weekly via Stripe Connect, based on hours logged
- Part-time (20 hrs/week) with fully remote, async flexibility — work from anywhere, on your own schedule
Apply Now