FDE vs Applied AI Engineer
The two titles increasingly overlap. Both ship production AI work against real customer surfaces — but they emphasize different sides of the same problem.
The short version
Forward Deployed Engineer. Customer-embedded engineer shipping integration code against the customer's data and workflow. The emphasis is the customer surface — what does this specific organization need this AI capability to do?
Applied AI Engineer. Engineer applying frontier model capabilities to production product surfaces. The emphasis is the model surface — how does this capability behave in production, and how do we evaluate and improve it?
At small companies and senior levels these titles collapse into one job. At larger orgs they remain meaningfully different roles.
| Dimension | Forward Deployed Engineer | Applied AI Engineer |
|---|---|---|
| Primary surface | Customer data, customer workflow, customer-specific integration | Product surface, model behavior, capability rollout |
| Code shape | Integration, ETL, custom workflows, end-to-end mission code | Prompts, evals, fine-tuning glue, capability-bound product code |
| Most common at | Palantir, OpenAI (Forward Deployed track), Scale AI | OpenAI (Applied AI), Anthropic (Applied AI Architect) |
| Research adjacency | Low — applied product work, not capability discovery | Moderate — sits next to research, often pairs with researchers |
| Career path forward | Senior FDE → Staff FDE → engineering or product leadership | Senior Applied AI → Staff Applied AI → research-engineering hybrid roles |
| Comp shape | Base + equity, no quota | Base + equity, no quota |
Frequently asked questions
Are FDE and Applied AI Engineer the same role?
At Anthropic, they often are — the Applied AI Architect title and the Forward Deployed Engineer title sit on the same team and target the same work shape. At OpenAI, they're close cousins: Applied AI Engineer leans toward product surface, FDE leans toward customer deployment. At Palantir, FDE / FDSE and Applied AI Engineer are distinct tracks with overlapping IC ladders.
Which role is closer to research?
Applied AI Engineer sits closer to model and capability research — work routinely touches fine-tuning, evals, prompt engineering, and production model behavior. FDE sits closer to product and integration — applying existing capabilities to a specific customer's data and workflow. The line blurs at the senior level.
Should I aim for FDE or Applied AI if I want to maximize learning?
Applied AI Engineer if your gap is model behavior, evals, and capability work. FDE if your gap is shipping integrations against messy real-world data and high-stakes customers. Both compound, but they compound in different directions: AAE toward model expertise, FDE toward customer expertise.
See live postings: Forward Deployed Engineer · Applied AI Engineer