3 days ago
New York, NY, USAMid Level / Senior
Base Salary
$220k - $320k/yr
Responsibilities
- Design and build production-grade RL environments from task specification through stable, versioned release.
- Write reward functions and programmatic verifiers while hardening environments against reward hacking and shortcut solutions.
- Build evaluation environments with seeded failure modes, model and agent harnesses, bounded behavior, and explicit output verification.
- Transform de-identified enterprise data into realistic agent state, tools, files, and context while maintaining reproducibility, observability, safety, and version-change evidence.
- Collaborate directly with frontier AI lab researchers to identify capability gaps and design task suites to address them.
- Run frontier models against environments to measure task difficulty and solvability, then tune environments to provide useful training signals.
- Set standards and reusable patterns for Replay's task and environment development.
Requirements
- 3-5+ years of ML or software engineering experience shipping production systems.
- Personally built production-grade RL environments used in real training or evaluation rather than only prototypes or research demos.
- Ability to translate ambiguous researcher goals into well-scoped working systems.
- Strong understanding of environment design tradeoffs including task specification, state and tool design, determinism, difficulty calibration, and solvability.
- Experience building evaluation or agentic harnesses and applying rigorous evaluation concepts such as precision, recall, calibration, imperfect labels, and distribution shift.
- High ownership, action-oriented approach with the ability to take problems from idea through shipment.
- Startup experience, broad ownership, and comfort with evolving requirements.
- Clear written and verbal communication with researchers, engineers, and non-technical partners.
- Bonus experience at or working with a frontier AI lab or in research-adjacent engineering.
- Bonus experience in forward-deployed or customer-facing engineering, sensitive enterprise data or other high-trust production systems, or as a founder or early startup employee.
Benefits
- $220-320K base salary.
- Meaningful equity.
- 100% covered medical, dental, and vision.
- Unlimited PTO.
- Free gym and workout class memberships through Wellhub.
- $500 desk allowance.
- Up to $15,000 relocation bonus where applicable.
Categories
About Sunset
Sunset provides software-enabled services that help tech companies wind down operations, handling state withdrawals, liquidations, and related compliance to reduce cost and complexity. It also offers tools that help businesses unlock new revenue streams beyond shutdowns. Founded in 2023 and headquartered in New York, the privately held company sells fee-based services and tooling to founders and operators seeking an efficient, compliant exit.