Forward-Deployed AI Data Engineer

Role Description

Most enterprise data environments were never built to be AI-ready. They were built to survive — cobbled together over years of acquisitions, migrations, and workarounds. The data exists. It's scattered, unlabeled, and structurally hostile to anything that assumes cleanliness.

You've worked in those environments. Not as an observer — as the person who had to make something work inside them. You know the difference between a schema that looks clean and one that is clean. You've hit the accuracy cliff with an LLM and built around it instead of pretending it wasn't there.

You're not looking for a greenfield project with perfect infrastructure. You're looking for the genuinely hard problem — and the chance to solve it in front of a customer who needs it solved.

What You'll Actually Do

• Lead technical onboarding and implementation from data environment discovery through production deployment
• Build, configure, and troubleshoot data connectors, pipelines, and AI agent workflows inside client environments
• Work directly with Forge, Lattice, and Stratum — our agent framework, orchestration layer, and semantic intelligence system
• Serve as the primary technical point of contact for your accounts post-deployment
• Surface what you're learning in the field — product gaps, failure modes, recurring patterns — back to engineering
• Develop implementation playbooks from each engagement so the next one goes faster
• Partner with the Enterprise Data Strategist and CEO on pre-sale scoping, technical discovery, and proof-of-concept builds

What Success Looks Like in Year One

• You've run multiple enterprise implementations end-to-end and have something running in production at each one.
• You've built playbooks from what you learned, not just completed the engagements.
• Clients are asking for you by name.
• The team trusts you to go in alone and come back with something that works.
• The measure isn't how clean the code was. It's whether the agents produced the right outputs, reliably, in an environment that was never designed for them.

Who We're Looking For

• 4–8 years combining hands-on data engineering with direct deployment or customer exposure — forward-deployed engineering, solutions engineering, data consulting, or technical implementation at a data or AI company
• Experience working inside enterprise data environments and familiarity with CRMs, warehouses, and legacy pipelines
• SQL fluency — you think in queries, use DuckDB, dbt, or similar without looking things up; proficiency in Python preferred; comfortable reading and writing API integrations
• Hands-on experience building or deploying AI agent workflows; you know where LLMs break against real data problems

The Stuff That's Harder to Teach

• Unstructured data instincts — no schema, no labels, no consistent format — and you didn't flinch.
• Bias toward output — you care more about whether the agent's results were right than whether the code was elegant.
• Client-facing comfort — you can sit in a room with a CTO and explain why their data isn't AI-ready without making them feel bad about it.
• Strong opinions — you have a clear view on why most AI deployments fail on data, not model — and you've built something that proved it.

Bonus (Genuinely Not Required)

• Experience at a company running a forward-deployed or consultative technical model — Palantir, Scale AI, or similar
• Familiarity with blockchain data, DeFi, or institutional crypto infrastructure
• Experience in financial services or insurance data environments

To Apply

Complete the online application and include responses to: 1) why this role fits where you are in your career right now, and why you are the right person for it; and 2) one example of a messy data problem you had to solve in production — what the environment looked like, what broke, and how you fixed it.

No template. Just tell us the story.