Senior ML Infra Engineer

Job description

**The Opportunity** The legal industry is broken — lawyers bill by the hour, charge enormous sums, and take forever to respond. This means that a lot of people who need legal advice don't get it. We're using AI to solve this problem. AI takes the first pass at everything our lawyers do, so that the lawyers can focus on human interaction. We're starting with negotiating commercial contracts, and will expand quickly from there to other use cases within legal. We're reimagining what a contract can be in the AI age. We're a veteran team, with one successful exit already, who've been doing deep learning in the legal space since long before ChatGPT. **Description** 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. **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 **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 **Compensation & Benefits** $200,000 - $275,000 base, calibrated to experience and location Health, dental, vision; unlimited PTO; wellness stipend & more…