Forward Deployed Engineering Tutorial · The Role · Lesson 3 of 24
Forward Deployed Engineering: Origins & Evolution
Where the forward deployed engineer role came from, how it grew from Palantir's "Deltas" into a common role at AI and enterprise software companies, and how it compares with older customer-facing engineering roles.
How the role started
Software vendors have always needed people to make products work for customers: professional services teams, implementation consultants, sales engineers. What made Palantir, founded in 2003, different was putting software engineers with the authority to write production code directly inside customer organisations, starting with government and later commercial customers. Palantir calls them Forward Deployed Software Engineers, internally "Deltas", alongside the "Devs" who build its platforms. In Palantir's words, Devs focus on a single capability for many customers while Deltas enable many capabilities for a single customer (Dev versus Delta).
Timeline
- AI and enterprise softwareCurrent
- Forward deployed engineering teams become common at AI companies and data platforms moving customers from demos to production
- Generative AI pilotsHistoric
- Companies find that impressive model demos need substantial integration, data and evaluation work to reach production, and start hiring FDEs
- Palantir goes publicHistoric
- Palantir lists on the New York Stock Exchange, bringing wider attention to its operating model
- Palantir foundedHistoric
- Engineers are embedded with customers to deploy and adapt its software: the origin of the forward deployed role
How the role evolved
| Professional services / implementation | Forward deployed engineering (Palantir model) | FDE in the AI era | |
|---|---|---|---|
| Main output | Configured product, delivered project | Production solutions for one customer, product feedback | Production AI workflows: integrations, retrieval, evals, guardrails |
| Code | Scripts and configuration | Substantial production code | Substantial production code plus AI evaluation |
| Product feedback | Limited | Core part of the role | Core part of the role, fast-moving products |
| Typical employers | Enterprise software vendors, consultancies | Data platforms, defence and government software | AI labs, AI application companies, data platforms |
What the role looks like today
- Engineering-first: FDEs are hired and levelled as engineers, with practical coding interviews and customer case rounds.
- AI-heavy: many openings focus on deploying language models: workflow design, retrieval, evaluation, safety and cost.
- Product leverage: strong FDE teams upstream what they build, so each deployment makes the product better for the next customer.
Resources
- Palantir: Dev versus Delta
- Palantir: A day in the life of a Forward Deployed Software Engineer
- NIST AI Risk Management Framework, a common reference when deploying AI in regulated customers
Last reviewed 21 October 2026. Release and support dates come from the official release pages linked above.