The FDE List
Live indexThe best Forward Deployed Engineer jobs, ranked.
FDE is tech’s fastest-growing job title. This is the living index of the programs that matter — real compensation, interview loops, and what the work is actually like, re-verified as the market moves. Every number traces to a posted range, a live careers page, or levels.fyi.
Sources: posted salary ranges, levels.fyi, and company careers pages. Last verified September 2, 2026. How we rank →
FDEs in OpenAI’s Model Deployment for Business org own frontier-model deployments end to end — discovery, scoping, full-stack builds on OpenAI APIs, and production rollout — for strategic enterprise and government customers. What they learn in the field feeds directly back into research and product.
Interview loop, score breakdown & sources — read the full OpenAI FDE guide →
Palantir invented the FDE model. FDSEs embed at customer sites building on Foundry, Gotham, and AIP for defense, intelligence, hospitals, manufacturers, and banks — with the paired Deployment Strategist track handling the operational side of the same deployments.
Interview loop, score breakdown & sources — read the full Palantir FDE guide →
Anthropic FDEs embed with strategic customers to ship production Claude applications — building agents and MCP servers, providing white-glove deployment support, and feeding repeatable patterns back to product. Roughly 40% full-stack build, 30% architecture, 30% discovery.
Interview loop, score breakdown & sources — read the full Anthropic FDE guide →
Google Cloud FDEs are “embedded builders” who code, debug, and ship bespoke agentic solutions inside customer environments, with direct access to DeepMind engineering. A separate DeepMind FDE track embeds with strategic partners on evals and frontier deployments.
Interview loop, score breakdown & sources — read the full Google FDE guide →
Databricks FDEs embed with customer teams to ship production data and AI systems — lakehouse migrations through first-of-a-kind GenAI apps (RAG, multi-agent, Text2SQL) built on Mosaic AI. Engagements are outcome-based with shared customer OKRs, not billable hours.
Interview loop, score breakdown & sources — read the full Databricks FDE guide →
Sierra’s forward-deployed function is its Agent Engineering org — reqs post as “Software Engineer, Agent”: engineers own the Agent Development Life Cycle end to end, building and tuning production customer-facing AI agents for brands in finance, healthcare, telecom, and commerce. Deployed Infrastructure Engineers install the platform into customer VPCs.
Interview loop, score breakdown & sources — read the full Sierra FDE guide →
AWS’s Forward Deployed Engineering org embeds teams of five to six engineers inside enterprise customers for sprints of 45 minutes to 45 days (the “AI 45” model), co-building production agentic AI on Bedrock, Kiro, and Amazon Q — designed to leave customers self-sufficient, with runbooks, knowledge graphs, and trained staff.
Interview loop, score breakdown & sources — read the full AWS FDE guide →
Salesforce FDE pods embed full-time with a single enterprise customer for roughly three-month engagements, personally writing the code that designs, builds, and ships custom Agentforce agents into production. The program spans associate level through principal.
Interview loop, score breakdown & sources — read the full Salesforce FDE guide →
Microsoft’s new Frontier Company is a dedicated subsidiary whose engineers sit inside client organizations to design, deploy, and run AI systems on-site. Posted FDE reqs build user-facing GenAI features end to end for major enterprise brands on the Copilot and Azure AI stack.
Interview loop, score breakdown & sources — read the full Microsoft FDE guide →
Perplexity’s FDEs — hired as Members of Technical Staff, Applied AI — design, build, and deploy end-to-end integrations that embed Perplexity into enterprise systems, owning deployments from architecture through production rollout and optimization.
Interview loop, score breakdown & sources — read the full Perplexity FDE guide →
Ramp’s forward-deployed team designs end-to-end AI solutions for its largest enterprise customers, working directly with customers alongside sales and core product engineering. It’s the Palantir playbook applied to fintech, now productized as Ramp’s Applied AI Solutions offering for enterprise finance teams.
Interview loop, score breakdown & sources — read the full Ramp FDE guide →
Scale FDEs build customer-specific data infrastructure on the GenAI Data Engine — RLHF and evals pipelines — for frontier labs and enterprises, while Public Sector FDSEs deploy the platform inside secure federal environments.
Interview loop, score breakdown & sources — read the full Scale AI FDE guide →
Cursor (Anysphere) FDEs embed with enterprise engineering teams to ship production workflows on Cursor itself — large-scale refactors, migration systems, automated PR-review loops, and spec-to-implementation pipelines. The org is hiring FDEs and FDE leadership across the US, Europe, and APAC.
Interview loop, score breakdown & sources — read the full Cursor FDE guide →
Cohere FDEs bridge the North agentic platform and enterprise engineering teams, integrating AI into finance, healthcare, telecom, and government with a security-first posture — including sovereign-AI and public-sector deployments.
Interview loop, score breakdown & sources — read the full Cohere FDE guide →
Mistral’s customer-facing Applied AI org deploys its frontier models into enterprise production — fine-tuning, agents, and LLM applications from pre-sales through implementation — with a distinctly European footprint spanning sovereign and critical-infrastructure customers.
Interview loop, score breakdown & sources — read the full Mistral FDE guide →
Snowflake’s forward-deployed function re-emerged in 2026 across two tracks: Applied AI FDEs who embed with customers on Cortex-era AI workloads, and specialist Migration and Spark FDEs who move enterprise estates onto Snowflake. Reqs are centered on Menlo Park with a Warsaw outpost.
Interview loop, score breakdown & sources — read the full Snowflake FDE guide →
C3 FDEs design integration architectures, write production code in customer environments, and act as the primary technical contact for enterprise accounts on the C3 Agentic AI Platform — industrials, energy, defense, and federal.
Interview loop, score breakdown & sources — read the full C3.ai FDE guide →
IBM FDEs work directly with clients and product teams to deploy strategic IBM Z transaction-processing software and watsonx AI tooling from proof-of-value to production — mainframe modernization meets forward deployment.
Interview loop, score breakdown & sources — read the full IBM FDE guide →
The comp landscape
Posted base-salary bands for each program’s core FDE role, US market — equity and bonus come on top, and at the frontier labs they can double the number. Bands span levels; senior hires land near the top. Full breakdown in the FDE salary guide.
Cursor, Mistral, IBM don’t post salary ranges and are omitted here.
What is a Forward Deployed Engineer?
A Forward Deployed Engineer (FDE) is a software engineer who works embedded with customers — shipping production code against real customer problems instead of a product roadmap. Palantir invented the role two decades ago; the AI wave made it the industry’s hottest job title, because getting LLMs to work inside a Fortune 500 is as much a deployment problem as a modeling one.
The job blends software engineering with consulting-grade customer skills: scoping an ambiguous business problem, building the integration or agent that solves it, and demoing to executives — often on-site. It pays like engineering, promotes like sales, and compounds career capital faster than either.
New to the role? Start with our guides: What is an FDE?, FDE vs. Solutions Architect, and the FDE resume guide.
How we rank
Every program gets an Index score out of 100, built from four weighted dimensions. Scores are editorial judgments grounded in the data below — not paid placements.
Posted salary bands, levels.fyi data, and verified offer reports — total comp where equity data exists, base bands otherwise.
How real the FDE org is: age, scale, training, leadership, and whether forward-deployed work sits at the core of the business.
What the role does for the rest of your career — brand strength, alumni outcomes, and the caliber of problems you get to claim.
Open req counts, growth trajectory, and strategic investment — is the program expanding or coasting?
We also track Meta, NVIDIA, and SpaceXAI (formerly xAI), currently excluded from the ranking: Meta began posting Enterprise AI FDE reqs in August 2026 but the program is too nascent to score, NVIDIA’s first FDE-titled req (Senior Staff, posted August 2026 at a $224k–$356.5k band) is a single opening inside a Solutions-Architect-led motion, and xAI’s lone FDE req disappeared after the SpaceX merger — no FDE-titled roles remain on any SpaceXAI board. Snowflake, excluded in July, rejoined the ranking when its FDE reqs reopened.
Compensation data comes from posted salary ranges (required in several US states), levels.fyi, and public offer reports; figures are US-market and vary by level and location. Hiring status reflects each company’s careers page as of September 2, 2026. Spotted something outdated? Email [email protected].
Frequently asked questions
Land the offer, not just the interview.
Practice FDE interview questions, mock interview with peers, and learn the frameworks hiring managers actually score you on.
