Frontier Coding Data Engineer — Evaluation & Training

Surge AI; United States; Remote; USD 140,000 - 210,000; Full time

Job summary

Surge AI seeks a Research Engineer focused on Coding Evaluation and Training Data to design, run, and optimize data-driven tasks that teach frontier models how to code. You will own projects end-to-end, crafting tasks, RL environments, and evaluation schemes that reflect real-world software engineering challenges.

You will work with client engineers and Surge teams to translate high-level goals into concrete data workflows, tools, and environments, ensuring high-quality results.

Qualifications

• 3–6+ years of professional software engineering experience.
• Strong coding ability in at least one mainstream language.
• High taste for correct, high-quality engineering and production code.
• Ability to reason about and debug technical environments including containers, dependencies, and automation.
• Interest in owning projects end-to-end from scoping to execution and improvement.
• Excellent communication skills to speak credibly with senior client engineers.

Responsibilities

• Own end-to-end coding data projects from scoping and pilot design through execution and scale-up.
• Design agentic training workflows and task structures reflecting real-world SWE work.
• Define rubrics, golden sets, and reward signals that capture engineering value.
• Evaluate data and worker output against technical standards.
• Design and run qualification processes for coding workers.
• Set up complex environments (containers, repos, test harnesses, sandboxes, code execution infra).
• Collaborate with Surge engineering, product, and operations to improve coding data products.

Skills

Software engineering

Production code

Code review

Debugging

Communication

AI/ML systems

Tools

Containers

Test harnesses

Code execution infra

Job description

Surge AI seeks a Research Engineer focused on Coding Evaluation and Training Data to design, run, and optimize data-driven tasks that teach frontier models how to code. You will own projects end-to-end, crafting tasks, RL environments, and evaluation schemes that reflect real-world software engineering challenges.

You will work with client engineers and Surge teams to translate high-level goals into concrete data workflows, tools, and environments, ensuring high-quality results.