3 days ago
Base Salary
$250k - $280k/yr
Responsibilities
- Build evaluation systems that run millions of agent trajectories to measure model and product quality.
- Develop fine-tuning pipelines that convert evaluation signals into measurable agent improvements.
- Create agent-first product experiences, including UX and infrastructure for model and agent-operator workflows.
- Build systems supporting AI interviews, worker sourcing and matching, and the throughput needs of frontier labs.
- Scale reliable production infrastructure and integrate newly released models and capabilities quickly.
- Set technical direction, make and defend architecture and product decisions, and provide reference-quality designs and code.
Requirements
- At least 4 years of experience shipping systems relied upon by customers and other engineers.
- Ability to rapidly build full-stack prototypes that can become durable production foundations.
- Strong system design, API design, architecture, and product judgment.
- Experience working across all parts of the software stack.
- Deep proficiency in TypeScript and/or Python.
- Ability to operate effectively in ambiguous, fast-moving environments and influence through technical example.
- Preferred: production experience with LLM- or agent-driven products, LLM and agent evaluations, high-quality ML data, distributed systems, ML infrastructure, or large-scale data systems.
Benefits
- High-impact, early-stage-startup environment with expanded responsibility and career growth tied to contributions.
- Autonomy, clear ownership, continuous learning, and opportunities to shape frontier AI infrastructure.
- Fast-paced, high-intensity environment with Fortune 500 and leading AI lab customers.
- United States-based role with an annual base salary range of $250,000–$280,000; equity and additional benefits may be provided separately.
Tech Stack
Categories
About Labelbox
Labelbox builds and operates reinforcement learning data factories for the world’s leading AI labs and enterprises, powering the next generation of frontier models and AI applications.