Engineering Manager, Agents — SF

At Harvey, we're transforming how legal and professional services operate by combining frontier agentic AI, an enterprise-grade platform, and deep domain expertise.

This is a rare chance to help build a generational company at a true inflection point. We have strong product-market fit and world-class investor support. We're scaling fast and defining a new category in real time. The work is ambitious, the bar is high, and the opportunity for growth , personal, professional, and financial , is unmatched.

Our team moves fast, takes ownership, and is deeply committed to the mission , operating with intensity, staying close to our customers, and pushing each other for excellence. We live by three values: Decisiveness, Simplicity, and Job's Not Finished.

Role Overview

Harvey is redefining what's possible in legal work by combining cutting-edge AI with intuitive product experiences. As we scale rapidly and take on increasingly complex use cases, we're expanding our product engineering team to accelerate our ability to ship fast, delight users, and build infrastructure that scales with us.

Agents at Harvey let lawyers turn their expertise into reusable, automated work: custom agents and workflows that draft, review and analyze across a firm's documents. The Agents team owns Agent Builder, the Agents experience, workflows and the foundations underneath them: orchestration, durable execution, permissions and reliability. Other Harvey products build on this platform as well.

As Engineering Manager, you'll lead a large, senior team of full-stack and backend engineers through the shift from single-turn assistants to long-running, multi-agent systems that firms trust with their most sensitive work. This role is equal parts product judgment and systems depth.

Why Now

• Agents are how firms put Harvey to work: lawyers are turning their expertise into reusable agents and workflows that run across their documents and matters.
• The work is moving from single-turn assistance to long-running, scheduled and multi-agent work. That demands durable execution, sandboxed tools and recovery from failure at scale.
• Agents now reach deep into a firm's most sensitive data, so permissions, tenant isolation and governance have to hold everywhere an agent can act.
• Other Harvey products are building on this platform, so its quality and speed set the pace for much of the company.
• We need a leader who can raise the bar on reliability and evaluation while the team keeps shipping quickly.

What You'll Do

• Lead the Agents engineering team, a group of senior full-stack and backend engineers building Agent Builder, the Agents experience, workflows and the runtime underneath them.
• Own reliability and quality for agent and workflow execution, including on-call health, incident follow-through, SLOs and evals as part of how the team ships.
• Set technical direction for orchestration, durable execution and agent composition, and stay close enough to the code to make sound architecture calls and unblock hard problems.
• Hold a high bar on security and customer data boundaries, so permissions, isolation and admin controls are designed in from the start rather than patched later.
• Partner with Product, Design and Data Science on what agents should do for lawyers. Work with Agent Platform, Infrastructure and Security on the shared foundations.
• Shape how the team builds with AI coding agents: review practices, verification and test infrastructure that keep quality high as output grows.
• Hire, develop and retain strong engineers, grow tech leads, and build a team culture of ownership and candor.

What You Have

• You have 3+ years managing engineering teams, including teams of 10 or more, and 5+ years as a backend or full-stack engineer before that.
• You've shipped LLM-powered or agentic products in production and understand orchestration, tool use, evals, and latency and cost trade-offs from experience.
• You've operated multi-tenant systems with strict security, permission or compliance requirements, and you treat a permissions bug as a trust issue.
• You're comfortable going deep on distributed systems: workflow engines such as Temporal, queues, sandboxing and long-running jobs.
• You have strong product judgment and can weigh speed, quality and user impact in a domain where accuracy matters.
• You lead by raising the bar on quality without slowing the team down. You know when to step in and when to let your tech leads lead.
• Nice to have: experience with builder, low-code or automation tools for non-engineers, or with legal or other professional-services workflows.

Compensation Range

$260,000 - $330,000 USD