Forward Deployed Engineer, Enterprise Document AI

On-site (San Francisco, CA) | $220K to $320K | 2 openings

PLEASE INCLUDE YOUR LINKEDIN TO BE CONSIDERED!

About the company

A Series B AI infrastructure company that turns messy, unstructured documents (PDFs, scans, spreadsheets, forms) into clean data that LLMs and downstream systems can actually use. Most enterprise data lives in formats like these, which makes this one of the most durable problems in enterprise AI.

The company has grown revenue several times over in the past year, serves leading AI-native companies and some of the largest enterprises in the world, and is backed by top-tier venture firms. It's still a team of under 100 people, founded by engineers with a research background.

The role

This is the first forward deployed hire, so you'll help define what the function looks like as it scales. You'll split your time between pre-sales, where you prove the product against competitors on a customer's own documents, and post-sales, where you own technical success for a small set of enterprise accounts.

Customer-facing here means building: eval apps, custom agents, and changes to the product itself, not answering support tickets. The team is hiring 2 engineers at once, so you'll be building out the function alongside a peer rather than alone. You'll work directly with the founders and the head of engineering.

What you'll do

• Run evals and build custom demo applications with the sales team to win head-to-head evaluations against other document AI vendors
• Own the technical relationship with 1 to 3 enterprise accounts, meeting with them weekly or every other week to clear blockers and grow usage into new teams and use cases
• Build whatever moves a customer forward: an eval tool, a tailored agent, or an extension to an existing product surface
• Create diagnostics, dashboards, and eval harnesses that measure parsing quality and catch regressions across customer environments
• Bring what you learn in the field back to product and engineering, with clear evidence of gaps and unmet needs
• Get to know each account's technical stakeholders well enough to spot expansion before it's asked for

What you must bring

• 3+ years of experience building and deploying production software, ideally for enterprise or regulated customers
• Experience at a high-growth startup shipping production software to enterprise customers
• Strong Python, plus comfort in Bash and infrastructure work when a deployment needs it
• Daily, real use of AI coding tools (Claude Code, Codex, or similar) well beyond autocomplete, and a working understanding of LLM and ML concepts
• A backend-leaning full-stack, product engineering, or forward deployed background
• Comfort leading technical conversations on customer calls
• The habit of building your own tools to test an idea, such as a quick Streamlit app or a purpose-built dataset
• A quantitative, fast iteration style, and a quality bar where "mostly right" isn't done
• Careful attention to detail. In document parsing, one wrong character can change a number that matters

Nice to have

• Experience founding a company or building products at a very early stage
• Prior forward deployed work where your effort drove a large customer's success
• Hands-on experience designing and running evals
• Strength on both the deep engineering side and the customer-facing side

Compensation & logistics

• $220K to $320K base salary plus highly competitive equity
• Benefits include unlimited PTO, daily lunch, reimbursed transportation, medical, dental, and vision coverage, a monthly wellness budget, and flexible parental leave
• On-site five days a week in San Francisco, CA
• 2 openings
• Open to visa transfers, including OPT and H-1B transfers

This isn't the right fit if

• Your experience is only at large, established tech companies with no time at a fast-moving startup
• You've been in solutions or sales engineering without writing and deploying code on your own
• Your AI tooling stops at autocomplete suggestions
• You'd rather avoid customer calls altogether
• Your recent history includes several stints of under a year