RL Deep Learning Engineer

Location: New York, NY (On-site)
Company Stage of Funding: Early-stage / Seed-backed
Office Type: On-site
Salary: $210,000 – $250,000 Equity (0–0.4%)
Visa: No sponsorship available

Company Description

Our client is building the foundational legal data infrastructure powering the next generation of AI systems.

Their platform processes and structures millions of U.S. court records and legal filings, serving as the legal data layer for AI labs, legal AI startups, and enterprise legal workflows. The company already powers legal reasoning workflows for hundreds of law firms and multiple large-scale AI organizations.

This is an opportunity to join a highly technical early-stage team building RL environments, evaluation harnesses, and benchmark systems for long-horizon legal reasoning tasks — infrastructure that will directly shape how future legal AI models are trained and evaluated.

The role sits at the intersection of reinforcement learning infrastructure, evaluation systems, data pipelines, and large-scale document processing.

What You Will Do

Build and maintain RL environment infrastructure for long-horizon legal reasoning tasks

Design scalable evaluation harnesses, task runners, scoring systems, and sandboxed execution environments

Build systems that convert raw legal filings and court records into benchmark and RL training tasks

Develop contamination-free evaluation pipelines for frontier AI model testing

Integrate with partner model APIs and evaluation harnesses

Collaborate with attorneys and domain experts to translate legal workflows into structured evaluation tasks

Work with messy, large-scale, real-world document datasets including PDFs and long-form legal filings

Build scalable data pipelines for legal reasoning environments

Develop tools for search, retrieval, reasoning, and drafting evaluations

Write production-quality Python systems with strong engineering rigor

Contribute to infrastructure supporting thousands of concurrent agent evaluations

Leverage AI coding tools such as Cursor, Claude Code, and Codex in daily workflows

Collaborate closely with engineers, attorneys, and AI partners

Operate with high ownership in a lean, fast-moving engineering environment

Take full ownership of systems from 0 → 1

Ideal Candidate Background

3–8 years of software engineering experience

Strong Python engineering fundamentals

Experience building production systems with high ownership

Experience building systems from 0 → 1

Comfortable working with large-scale document or data processing systems

Strong backend and infrastructure engineering intuition

Experience working in high-signal startup or engineering environments

Strong product ownership mindset

Comfortable operating as a highly autonomous IC

Experience with AI coding tools in daily workflow

Strong debugging and systems thinking ability

Comfortable working with ambiguous and evolving requirements

Strong written and verbal communication skills

Ability to move quickly while maintaining engineering quality

Comfortable working onsite in NYC

Strong Signals

Founding engineer or startup founder experience with demonstrated traction

Experience building evaluation systems or benchmarking infrastructure

Experience with RL environments or agent evaluation systems

Experience with LLM evaluations or AI evaluation frameworks

Experience with modern AI tooling and workflows

Experience building scalable Python backend systems

Experience with large-scale data or document pipelines

TypeScript experience

Experience working with messy real-world datasets

Strong engineering ownership and initiative

Experience collaborating with highly technical teams

High agency and startup intensity tolerance

Evidence of rapid career growth or exceptional ownership

Strong systems design and infrastructure intuition

Compensation and Benefits

Base salary: $210,000 – $250,000

Equity package up to 0.4%

Direct ownership over core AI infrastructure systems

High-impact role at an early-stage AI infrastructure company

Work directly with frontier AI labs and legal domain experts

Lean, highly technical engineering team

Exposure to RL systems, evaluation harnesses, and large-scale AI infrastructure

Fast-growing company with strong commercial traction

Opportunity to shape the future of legal AI evaluation systems

Why Join

This is an opportunity to build foundational RL infrastructure and evaluation systems for frontier legal AI.

You’ll work on difficult, high-leverage engineering problems involving long-horizon reasoning, document intelligence, scalable evaluation environments, and AI benchmarking systems.

If you want deep technical ownership, exposure to frontier AI infrastructure, and the opportunity to help define how future AI systems are evaluated and trained, this role offers exceptional scope and leverage.