RL Deep Learning Engineer (Remote)

Ovii's Interpretation of the Role

Midpage seeks a remote RL Deep Learning Engineer to design, build, and scale evaluation harnesses and data pipelines that turn millions of court filings into reinforcement‑learning benchmarks for frontier AI labs.

Role Snapshot

• Build RL evaluation environments
• Own legal‑data pipeline
• Scale task runners for thousands of agents
• Integrate with model APIs
• Collaborate with attorneys and AI partners
• Full‑stack ownership
• Remote, full‑time

Must-Have Requirements

• Python
• software engineering fundamentals
• building and scaling production systems

Nice-to-Have Signals

• TypeScript
• Familiarity with LLM evaluation
• Comfort with messy legal document data
• LLM evaluation
• legal document handling

Work Setup

• Location: Global
• Work mode: REMOTE
• Remote scope: GLOBAL
• Employment type: Full-Time

Not Specified in JD

• Salary range
• Visa sponsorship
• Remote eligibility specifics
• Education requirement
• Certifications
• Relocation
• Notice period
• Travel
• Security clearance
• Coding test
• Portfolio
• GitHub
• Cover letter

What You'll Likely Work On

• Design and maintain a scalable evaluation harness and sandboxed RL environment infrastructure that can run thousands of parallel agents.
• Own the end‑to‑end data pipeline that transforms fresh court filings into benchmark and RL training tasks.
• Integrate partner harnesses and model APIs to enable contamination‑free evaluations.
• Translate legal workflows (cite checks, motion drafting, precedent research) into structured, scorable task formats using the Harbor spec.
• Collaborate directly with attorneys and frontier AI labs to ensure the environments meet real‑world legal reasoning needs.

Good Fit If You Have

• Self‑managing and kind attitude
• Clear written and verbal communication
• Interest in LLM evaluation and benchmark leakage
• Comfort handling messy legal documents (PDFs, long‑form text)