Senior Software & AI Engineers — Coding Benchmarks & RL Environments
Arc is looking to connect experienced Software and AI Engineers with a remote client project focused on the design and verification of programming tasks used to train and evaluate AI coding agents.
This opportunity is particularly relevant for engineers with experience in coding benchmarks, evaluation harnesses, reinforcement learning environments, or agentic software-engineering systems.
The work involves creating and validating high-quality programming tasks, including issue descriptions, test suites, and evaluation criteria used in AI training and assessment environments.
PLEASE NOTE - This is a platform-based opportunity. This is not a traditional-hire role.
Relevant backgrounds include
Backend, full-stack or frontend software engineering
ML / AI engineering
Coding benchmark or evaluation design
RL or agentic environments for software-engineering systems
Candidates should have
5+ years of paid, full-time-equivalent production software engineering experience
Current primary work in software engineering, ML / AI engineering, or coding-task authoring for AI systems
Direct experience authoring coding tasks with issue descriptions and test suites for AI training or evaluation
Experience building tasks or benchmarks used to train or evaluate software-engineering agents in an RL or agentic environment
Fluent English
Availability for at least 10 hours per week, starting next week
Location: Remote
Commitment: 10–20+ hours per week
Age requirement: 18+
If your background is a strong fit, feel free to apply.