Founding Forward Deployed Engineer

Our client | San Francisco, hybrid | Full-time | $180,000 to $230,000 base + equity

• You have 3 to 7 years in a customer-facing technical role (forward deployed, solutions, implementation or customer-facing engineering).
• You personally owned customer deployments end to end.
• You have worked at a venture-backed startup in an early-stage, zero-to-one setting.
• You have built and deployed LLM agents with customers, including prompts and evals.
• You write Python with FastAPI, and React.
• You are in the San Francisco Bay Area or California, or will relocate for a hybrid role.
• You are authorized to work in the US without visa sponsorship, now or in the future.

If one of these does not describe you, this role is probably not the right fit.

Our client is a customer command center built for the AI era. Older customer success tools such as Gainsight and Planhat give teams a dashboard and ask them to type data into it. Its AI agents do the work themselves: chasing renewals, running onboarding, and building quarterly business reviews.

Thousands of agents are live in production for enterprise customers and fast-growing AI companies. The company is about a year and a half old and still small. The founding team built generative AI and automation platforms at well-known technology companies, and the CTO is a third-time founder who has scaled engineering organizations past 100 engineers. The company is backed by leading venture investors, with angels from well-known AI and software companies.

For the first stretch, one engineering team built the platform and worked directly with customers. That team has now done many of those reps, and the platform handles most of what customers need. The bottleneck has moved from building the platform to deploying it: getting each customer live, teaching them how it works, and setting their agents up against their real systems.

You will be the first person to own that function. You are technical and good with customers, and you know how agents work, how to write prompts, and how to measure whether an agent is good enough to ship.

• Own a set of customer accounts through the six-week deployment cycle, from kickoff to go-live.
• Embed with customers to understand how they run onboarding, adoption and renewals, then identify where the agents add the most value.
• Connect the platform to each customer's systems, including Big Query, Salesforce and data warehouses. Set up SSO and integrations, and build the data pipelines.
• Build and deploy customer agents: gather requirements, write the prompts, and configure the agents against real customer data.
• Build evals that prove agent quality before you ship, then iterate until the agents clear the bar.
• Run customer meetings for roughly half your day or more, working alongside a product and deployment strategy counterpart who owns the relationship while you own the technical implementation.
• Partner with go-to-market to close, activate and expand accounts. You are the technical credibility in the room.
• Debug live deployments and run root cause analysis when an agent misbehaves in production.
• Work with the platform engineers when a customer needs a capability we do not have. For small fixes, write the ticket and the pull request yourself.
• Feed what you learn in the field back into the product. When the platform gets in the way, fix the platform.

Each account moves through a six-week motion split into a few phases. Depending on where an account is, you might be setting up integrations, getting data pipelines running, working through requirements, or building and tuning agents. Most deployments run virtually. Customers are mostly in California, with a growing group on the East Coast, and in-person visits happen occasionally but are not the main mode.

A normal day…