Senior/Staff Software Engineer, AI Systems

Hybrid role (3 days onsite) in either San Francisco or New York

Comp: $200-300k, competitive equity and strong benefits

Sponsorship available and open to visa transfers (e.g. OPT, H1B transfers)

Interviews: 3 rounds, all virtual and beginning with a recruiter call

My client is a leading Legal AI company based in San Francisco and New York.

I am looking for senior or staff-level software engineers to build AI systems that power complex professional workflows: drafting documents, reviewing evidence, and moving cases toward completion.

You'll own the agents and infrastructure behind these workflows — orchestration, backend services, and evaluations that measure quality and catch regressions — working directly with domain experts to turn early experiments into reliable production systems.

What you'll do

• Build agentic systems automating multi-step workflows, with human review at key points
• Redesign task/workflow orchestration for long-running work, dependencies, retries, and recovery
• Build evaluation systems to measure quality, catch regressions, and guide improvements
• Improve backend services, event pipelines, and scheduled job infrastructure
• Make agent execution observable for debugging, quality, latency, and cost management
• Connect systems to internal tools and a user-facing portal, including supervision APIs/interfaces

You might be a good fit if you:

• Have 4+ years shipping production software, with deep backend and AI/ML application experience in a fast-growing startup
• Have built with LLMs beyond basic chat, with real thought on tool use, context, and handling unreliable outputs
• Can translate domain experts' nuanced judgment into software behavior and evaluations
• Fluent with Python and TypeScript
• Have led projects end-to-end, from ambiguous requirements to deployment and operation
• Understand workflow orchestration and async systems (queues, events, retries, idempotency)
• Have high agency to research, prototype, and drive solutions independently

Experience with production agents, eval infrastructure, or document workflows is especially relevant. Engineering judgment matters more than familiarity with a specific framework.

Culture: Engineering works closely with ops, product, and customers; async-first with protected focus time; strong craft standards; small, tight-knit team.