Senior ML Infra Engineer

General Legal•San Francisco, CA

2d•$200,000 - $275,000•Hybrid

About The Position

General Legal is seeking a Senior ML Infra Engineer to build the systems that allow us to rapidly experiment with, evaluate, train, and deploy increasingly capable AI systems. You'll sit at the intersection of research and production engineering. Your job will be to make our researchers and engineers dramatically more effective: building reliable infrastructure for model experimentation, evaluation, inference, data generation, training, and observability while ensuring that promising ideas can move quickly from an experiment into production. This is a unique opportunity to join us at the ground level and define the AI infrastructure behind a platform that will set new standards for the legal industry. You'll have substantial autonomy over architecture and tooling and will help determine how our ML stack evolves as we scale.

Requirements

5+ years of software engineering, machine learning engineering, or infrastructure experience
Strong software engineering fundamentals and proficiency in Python
Experience building production infrastructure for machine learning systems
Experience with cloud infrastructure, distributed systems, and containerized workloads
Ability to independently design, build, and operate complex technical systems
Computer Science degree or equivalent experience
Eligible to work in the US, and able to work in-office a minimum of 3 days a week in our SF or NY office locations.

Nice To Haves

Experience training, fine-tuning, or serving large language models
Experience with GPU infrastructure and distributed training or inference
Experience with reinforcement learning, post-training, or large-scale evaluation systems
Experience building infrastructure for AI agents or long-running model workloads
Experience optimizing inference latency and cost at scale
Experience at an early-stage startup

Responsibilities

Build and own infrastructure for training, evaluating, and serving AI models
Develop systems for running large-scale experiments and evaluations quickly and reproducibly
Build pipelines for generating, processing, versioning, and managing training and evaluation data
Improve the reliability, latency, throughput, and cost efficiency of model inference
Build observability and monitoring for AI systems in production
Develop infrastructure for agentic workloads, including long-running and asynchronous model execution
Work closely with research scientists and product engineers to turn new AI capabilities into reliable production systems
Evaluate and integrate new models, inference systems, training frameworks, and infrastructure as the field evolves
Be proactive and come up with ideas on how to build our AI systems better