Forward Deployed Engineer

Ersilia; San Francisco (CA); On-site; USD 140,000 - 210,000; Full time

Job summary

Ersilia in San Francisco is seeking a Forward Deployed Engineer to own enterprise deployments end to end, from requirements discovery through live production. You will serve as the single accountable owner for each account, working directly with customers across government, healthcare, and enterprise to stand up the agent pipeline.

The role involves deploying on customer infrastructure, including bare-metal GPU environments, Docker containers, and Kubernetes orchestration, and requires strong

Qualifications

• Ability to deploy open-source models and debug production systems end to end.
• Experience engaging with government, hospital and enterprise stakeholders.
• Proven track record delivering client deployments.
• Location: SF office; relocation may be required.
• Schedule: five days in‑office; no 996.

Responsibilities

• Own full lifecycle of enterprise accounts from discovery to deployment.
• Deploy platform onto customer infrastructure including GPUs, Docker, Kubernetes.
• Gather requirements, create sample data, build evals and benchmarks.
• Lead mock-spec sessions with government and enterprise customers.
• Manage multiple accounts and escalate as needed.
• Handoff mature accounts and scale deployment capacity.

Skills

Technical deployment

Customer communication

Organization

Experience 0-6 years

In-person SF

In-office schedule

Tools

Docker

Kubernetes

Job description

Job Description

About Us

We specialise in building enterprise AI agent platform that deploys custom agentic workflows across highly regulated and complex environments, including governments, hospitals, and banks. The platform sits on proprietary primitives, including a hill-climbing algorithm that optimizes agents in context space, and handles everything from Slack integrations and routing to sandboxed infrastructure for regulated deployments. Customers include organizations across the public sector and large enterprises such as NBC, Toyota, and Domino's.

About the Role

We are hiring a Forward Deployed Engineer to own enterprise deployments end to end, from requirements discovery through live production. You will be the single person accountable for each account, sitting alongside customers across government, healthcare, and enterprise to understand their needs, stand up the agent pipeline, and iterate until the outcome is provably working. The work spans everything from bare-metal GPU deployments of open-source models under Docker and Kubernetes to running mock-spec sessions with ministry teams and building evals against customer-defined success criteria. This is a founding deployment hire and will report directly into engineering.

What You'll Own

• Own the full lifecycle of each enterprise account, from initial requirements discovery and customer calls through agent pipeline setup, deployment, and iteration
• Deploy our platform onto customer infrastructure, including bare-metal GPU environments, open-source model fine-tuning, Docker containers, and Kubernetes orchestration for regulated environments
• Gather requirements, generate sample data, build evals and benchmarks, and define success criteria in direct collaboration with customer stakeholders
• Run mock-spec and requirements sessions with enterprise and government customers, asking the right questions to surface what a working deployment actually looks like
• Manage three to four active accounts simultaneously once ramped, looping in core engineering as needed while remaining the single accountable owner for each
• Hand off mature accounts and expand scope as the customer grows, helping us scale its deployment capacity

Must-Have

• Strong technical foundation: able to deploy open-source models, work with Docker and Kubernetes, and debug production systems end to end without hand-holding
• Excellent customer-facing communication skills: comfortable owning calls with government, hospital, and enterprise stakeholders, setting requirements, and running structured spec sessions
• Highly resourceful and organized: able to run multiple accounts in parallel, drive each to a clear outcome, and escalte precisely when needed
• 0 to 6 years of experience in a deployment, forward-deployed, or technically customer-facing engineering role, or equivalent demonstrated ownership
• In-person at the San Francisco office; candidates unable to relocate may be considered only in exceptional circumstances
• Willingness to work the stated schedule: Five days a week in-office; intense, high-ownership pace with milestone-driven expectations. No 996 requirement.

Nice-to-Have

• Prior forward-deployed engineering experience at Palantir, Decagon, technical Sierra, Distill, C3, or a company with a similar FDE motion
• Technical consultancy background: has sold, scoped, and delivered client engagements, particularly with a technical component
• Entrepreneurial side projects, a small consultancy, or any experience getting in front of customers and solving problems under resource constraints
• Exposure to bare-metal GPU infrastructure, open-source model deployment, or fine-tuning in a production context
• Experience working with customers outside the United States, given the international composition of our customer base