Research Engineer — RL Training Data & Evals

HUD; San Francisco (CA); On-site; USD 120,000 - 180,000; Full time

Benefits offered by this job

Blue Shield medical, dental, vision

Lunch and dinner in office

Holiday break

Equinox membership

401k

Commuter benefits

Unlimited access to AI tokens

Job summary

HUD is actively building infrastructure to create RL training data and evals for frontier AI agents, with a focus on a broad application. The role is a general application for candidates interested in QC Automation, Benchmarks, or Synthetic Data, with visa sponsorship for strong US candidates.

This is a full-time on-site position in the San Francisco Bay Area. You will build systems, design experiments, and develop tools across the data lifecycle, collaborating with researchers, engineers, and

Qualifications

• Proficiency in Python and Linux environments.
• Experience with benchmarks and evals; assess task realism and data quality.
• Attention to data inconsistencies in models or tasks.
• Experience building tools, pipelines, experiments, or infra.
• Startup experience; able to work independently in fast-paced settings.

Responsibilities

• Build systems for creating, running, evaluating, and improving agent training environments
• Design experiments to understand model behavior, failure modes, and data quality
• Develop tools to help researchers, engineers, and data vendors create better tasks, trajectories, and feedback loops
• Work across the full lifecycle of agent training data from task design to evaluation
• Partner with external vendors to improve data engine quality and throughput
• Build metrics to assess usefulness of tasks, environments, and evals

Skills

Python

Linux

Attention to detail

Independent work

Tools

Docker

Data pipelines

Experimentation infra

Job description

HUD is actively building infrastructure to create RL training data and evals for frontier AI agents, with a focus on a broad application. The role is a general application for candidates interested in QC Automation, Benchmarks, or Synthetic Data, with visa sponsorship for strong US candidates.

This is a full-time on-site position in the San Francisco Bay Area. You will build systems, design experiments, and develop tools across the data lifecycle, collaborating with researchers, engineers, and