Forward Deployed Engineer

About the Role

We are hiring a Forward Deployed Engineer to work directly with customers on regulated, operational workflows and turn that field work into a production product.

FDEs sit close to the customer and close to the code. You will map real workflows, identify the first useful product slice, implement it, verify it, demo it truthfully, and help decide what should become a reusable platform. The work spans three modes:

• Field: shadow operators, model decisions, find evidence sources, and understand where the workflow is slow, risky, or brittle.

• Build: ship production slices across UI, API, data, permissions, tests, and deployment paths.

• Productize: turn customer-specific learning into primitives, configuration, evals, playbooks, or roadmap changes.

What You'll Do

• Run customer discovery, workflow shadowing, and field notes with operators and decision owners, until you can name the users, states, evidence sources, exceptions, and the real operating constraint

• Implement thin product slices in a live codebase across frontend, backend, data, and integrations, where the happy path works, unsafe paths fail, and the behavior survives realistic data

• Write tests, run realistic paths, inspect logs, and document what is proven, missing, or uncertain, so every customer demo is backed by evidence, not optimism

• Explain tradeoffs, risks, and next steps to non-technical customers without overclaiming

• Identify reusable patterns from field work and feed them into product and engineering, so the next customer gets faster onboarding, safer workflows, or more reusable product

• Carry ambiguous work end to end: discovery, build, demo, rollout, and follow-up

What We're Looking For

We don't need you to have used every tool in our stack. We need someone who can move safely across this kind of system and learn the missing pieces quickly.

• You have personally shipped production software and can explain what you touched, how you verified it, and what changed for users

• You are comfortable in messy customer settings where the first request is rarely the real problem

• You can talk to operators in plain language, then go back to the codebase and build the thing

• Product UI: TypeScript, React, Next.js, shadcn/Tailwind; you can trace a user flow, change a screen, and respect server/client boundaries

• Backend: Python (uv), Pydantic, Django/Django Ninja or FastAPI, background workers, and typed APIs

• Data and auth: Postgres, migrations, service roles, tenant scoping, and auditability

• AI systems: pydantic-ai agents, typed outputs, evals, and provider choice across Gemini, OpenAI, and Bedrock; you use AI tools for leverage but never treat generated output or a clean demo as proof

• Cloud: Vercel, Cloudflare, AWS, GCP; you can debug across deployment, env config, logs, and customer-facing behavior

• Evidence discipline: you naturally separate fact, inference, assumption, and risk

• Product judgment: you resist one-off customization unless the lesson clearly belongs outside core product

• Strong communication skills: you can explain what is safe, what is uncertain, and what happens next

Nice to Have

• AWS experience (IAM, GitHub OIDC, Lambda, API Gateway, Secrets Manager, CloudWatch, least privilege)

• Infrastructure as code experience

• Experience in regulated industries (energy, compliance, government permitting, healthcare, finance)

• Prior forward deployed, solutions, or founding engineer experience