Legal Operations
Cognition AI
- Location
- San Francisco
- Employment
- Full Time
- Work model
- On-Site
- Level
- Mid
- Posted
- 1h ago
About this role
We are an applied AI lab building end-to-end software agents. We're the makers of Devin, the first AI software engineer. Our team is extremely talent-dense. Among our founding team, we have world-class competitive programmers, former founders, and leaders from companies at the cutting edge of AI including Scale AI, Palantir, Cursor, Waymo, Tesla, Lunchclub, Modal, Google DeepMind, and Nuro. Building Devin is just the first step—our hardest challenges still lie ahead. If you’re excited to solve some of the world’s biggest problems and build AI that can reason on real-world tasks, apply to join us.
About The Role
Cognition is looking for an exceptional operator to help build an AI-native legal function for an AI-native company. This is not a traditional legal operations role. You will design how Legal operates: how work comes in, how decisions get made, how knowledge is captured, where AI and automation can replace repetitive work, and how a small team can support a rapidly growing company without recreating the large, siloed legal departments of the past. You will own the systems, technology, budget, outside counsel relationships, and operating processes that allow the Legal team to move quickly while maintaining extremely high standards. The goal is simple: build a legal organization that operates with the speed, leverage, and ambition of Cognition itself.
What You'll Do
Build an AI-Native Legal Function Design and continuously improve the operating model for Legal. Identify work that should be handled by lawyers, work that should be automated, and work that should disappear altogether. Build intelligent intake, triage, workflow, and knowledge systems that allow routine legal questions to be resolved quickly and consistently while directing the hardest problems to the right people. Work with our technical teams to experiment aggressively with Devin, AI agents, and other tools that can materially change how legal work gets done. We don't want to add AI to traditional legal processes. We want to rethink the processes themselves. Make Legal Easy to Work With Build systems that make it easy for people across Cognition to get good legal answers quickly. Work directly with Sales, Product, Engineering, Security, Finance, People, Communications, and other teams to identify friction and remove it. Create scalable processes without creating bureaucracy. The best legal process is often the one people barely notice. Measure what matters—speed, outcomes, recurring sources of friction, outside spend, and where the team's time is actually going—and use that information to continuously improve how we operate. Own Legal Technology and Automation Own the technology strategy for the Legal team, including contracting, knowledge management, matter management, e-billing, entity management, and other systems where appropriate. But don't assume we need traditional legal technology simply because other legal departments use it. Start with the problem. Determine whether software, AI, automation, a better process—or eliminating the process entirely—is the best answer. Build and deploy AI-enabled workflows that meaningfully increase the leverage of every person on the Legal team. Own the Business of Legal Own the Legal team's budget, outside counsel relationships, vendors, technology, and operational planning. Develop a deliberate approach to when we use outside counsel and when we build expertise internally. Negotiate fees and alternative arrangements aggressively and make sure external lawyers are delivering expertise rather than work we can do better ourselves. Build simple, useful operating rhythms and dashboards that give us a clear view of the team's priorities, workload, spending, and performance. Help Build the Legal Team Help us build a Legal team where exceptional people have unusually high levels of ownership. We expect people here to understand the business, exercise judgment, solve problems outside narrow job descriptions, and