Remote AI Infrastructure & Product Engineer

Irys, Legal AI; United States; On-site; USD 180,000 - 260,000; Full time

Job summary

Irys, Legal AI seeks an AI Infrastructure / Product Engineer to build evaluations, services, and product features that advance applied AI in legal work.

This full-time, remote role within the United States offers significant ownership on a distributed team, with direct access to leadership to shape measurement and quality improvements.

You will work on benchmarks, evaluation pipelines, dashboards, and production features across retrieval, drafting, document, and agent workflows.

Qualifications

• Strong Python skills and cloud deployment experience.
• Experience with APIs, databases, backend services, and deployment pipelines.
• Interest in LLMs, retrieval-augmented generation, and evaluation.

Responsibilities

• Build repeatable benchmarks for models, agents, and legal workflows.
• Develop evaluation pipelines for accuracy, hallucinations, retrieval quality, latency, and cost.
• Create dashboards to visualize system performance changes.
• Develop production features turning research prototypes into usable retrieval, drafting, document, and agent workflows.
• Implement backend services, APIs, and internal tools for testing and releasing improvements.
• Manage cloud deployments with monitoring and CI/CD pipelines.

Skills

Python

Cloud deployment

APIs & databases

Backend services

Deployment pipelines

LLMs / RAG

Full-stack experience

FastAPI

Docker

React/Next.js

Postgres

Vector databases

Tools

FastAPI

Docker

React/Next.js

Postgres

Vector databases

Job description

Irys, Legal AI seeks an AI Infrastructure / Product Engineer to build evaluations, services, and product features that advance applied AI in legal work.

This full-time, remote role within the United States offers significant ownership on a distributed team, with direct access to leadership to shape measurement and quality improvements.

You will work on benchmarks, evaluation pipelines, dashboards, and production features across retrieval, drafting, document, and agent workflows.