Competitive Programming RL Environment Developer

About OpenTrain

OpenTrain is the #1 platform for finding and building careers in AI training and data labeling. We help people start and grow freelance careers teaching AI by bringing specialized AI-training work and tools together in one place.

Why AI training work matters

AI training (also called data labeling or human feedback work) is the human side of building modern AI: people prepare, evaluate, and guide model behavior using real examples. This work is highly flexible, often remote, and a fast-growing route into tech where contributors shape how state-of-the-art systems behave.

• Flexible, remote work that fits around other commitments
• Accessible entry points — many projects need domain fluency or attention to detail more than prior experience
• Direct impact on how AI models solve problems and reason about code

The role

OpenTrain is hiring a Competitive Programming RL Environment Developer to create reinforcement-learning environments that evaluate AI models on original competitive programming problems. You will design problem checkers that accept multiple valid outputs, produce reference solutions, and author robust test suites and documentation used for fine-tuning and evaluation.

• Position type: Contractor, part-time
• Location: Remote — worldwide (English required)
• Time commitment: 20+ hours/week
• Pay: Hourly, $45–$65 USD per hour

What you'll do

This role focuses on engineering reliable, efficient infrastructure that judges algorithmic solutions produced by models. Work will include developing C++ checkers, validating submissions, and maintaining clear documentation so reviewers and problem setters understand edge cases and limitations.

• Design and implement checkers for problems with multiple valid outputs and nuanced acceptance rules
• Create reference solutions and comprehensive test suites that exercise edge cases and performance constraints
• Evaluate and validate participant submissions against complex problem statements drawn from competitive programming-style sources
• Write maintainable C++ code focused on correctness, performance, and clarity
• Document checker logic and explain tricky edge cases in writing and verbally
• Provide actionable feedback to problem setters and reviewers to improve test coverage and checker robustness
• Keep up with competitive programming standards and incorporate best practices into checkers and tests

Requirements

You must be comfortable working independently in a remote, collaborative setting and delivering well-documented, production-quality code under deadlines.

• Expertise in competitive programming and coding problem analysis
• Advanced proficiency in C++ for implementing robust checker solutions
• Experience designing and implementing problem checkers that accept multiple valid outputs
• Strong analytical skills to interpret complex constraints and multiple solution paths
• Exceptional written and verbal communication to articulate technical details clearly
• Meticulous attention to detail for code validation and output consistency
• Ability to produce high-quality, well-documented code on tight timelines

Helpful background

The following experience is not required but will make your application stand out and help you ramp faster:

• Prior experience as a problem setter, tester, or checker on major competitive programming platforms
• Participation as a competitor, judge, or organizer in competitive C++ events (e.g., national/international contests)

How this work is used & logistics

Your checkers, reference solutions, and test suites will support fine-tuning and evaluation pipelines (labeling types: FINE_TUNING and EVALUATION_RATING) focused on computer code/programming data. OpenTrain is the contracting organization for this role and will coordinate assignments, review, and payment.

To apply, be prepared to share examples of C++ code you wrote for algorithmic problems, and to discuss how you would handle ambiguous outputs, floating-point tolerance, and performance constraints.

• Data type: Computer code / programming
• Labeling focus: Fine-tuning and evaluation/rating
• Languages: English required
• Experience level: Entry level (applicants new to the industry but with the required skills are welcome)