Role Description
Location: San Francisco (In-Person) OR New York (In-Person, with an initial 3–6 month rotation in SF)
About the Role:
We are looking for a Forward Deployed Engineer (FDE) who builds like a core systems engineer and communicates like a peer to elite researchers. You will embed directly with technical teams at customer organizations—working shoulder-to-shoulder with their machine learning engineers, researchers, and infrastructure leads—to architect, code, and deploy complex AI agent systems into production.
This is not a traditional solutions engineering or technical account management role. You will write high-volume, production-grade code, solve non-trivial architecture challenges in real time, and push the limits of what automated agent workflows can accomplish in enterprise environments.
What You'll Do
- Code & Ship Systems: Design, build, and deploy production-ready AI agent architectures tailored to high-complexity customer environments (think Decagon, Sierra, Harvey, or Glean-style workflows).
- Engage Technical Peers: Lead deep technical discussions, code reviews, and architecture reviews directly with client engineers and ML researchers, establishing immediate credibility.
- Bridge Product & Customer: Extract core engineering patterns from client deployments to inform and directly contribute to our core platform and product roadmap.
- Unblock Deployment Pipelines: Diagnose performance bottlenecks, hallucination vectors, state-management failures, and latency issues across high-throughput agent pipelines.
What We're Looking For
- Agent Architecture Experience: Proven track record building complex agentic systems, tool-use frameworks, autonomous workflows, or high-density retrieval/reasoning engines.
- Hands-on Technical FDE Track Record: Prior experience in a forward-deployed engineering role where you explicitly wrote production code, or an engineering role with a heavy customer-facing edge.
- Stack-Agnostic Software Fundamentals: Strong foundation in computer science, system design, data structures, and production backend code—regardless of specific language or framework.
- Engineering Credibility: Ability to go white-board-to-white-board on LLM internals, evaluation frameworks, memory management, and edge cases with top-tier AI researchers and engineers.
Location & Mobility
- San Francisco: In-person full-time OR
- New York: In-person full-time, with an initial 3 to 6-month immersion/rotation in San Francisco before settling in NYC.
About Forward Deployed Engineering
Forward Deployed Engineers are embedded directly with customers to build custom solutions, integrate products into existing infrastructure, and bridge the gap between product engineering and customer success. The role combines deep technical skills with the ability to operate in client environments and translate business requirements into working software.
Originally pioneered by Palantir, the FDE model has spread across AI, enterprise SaaS, and cloud infrastructure companies. FDEs write production code, architect integrations, train customer teams, and feed product insights back to the core engineering organization. At companies like OpenAI, Salesforce, and Databricks, FDE teams are treated as elite engineering units that can ship custom solutions in days rather than quarters.
Typical FDE stack: Python, TypeScript, SQL, REST/GraphQL APIs, cloud platforms (AWS/GCP/Azure), and increasingly LLM APIs and AI orchestration frameworks. Strong communication and the ability to context-switch between technical and business conversations are as important as coding ability.
Similar Roles
Get the FDE Pulse Brief
Weekly market intelligence for Forward Deployed Engineers. Job trends, salary data, and who's hiring. Free.