What is a forward deployed engineer? The job, the pay and who hires

Everything below comes from Palantir's blog and S-1 filing and from the employers' own announcements, blogs and FDE postings, read on 2 October 2026. Where I quote a newsletter or an investor's opinion, I label it.

Where the title comes from: Palantir's Deltas

In 2019 Palantir's blog named its two biggest engineering roles by headcount, each with an internal name. Devs, the software engineers, build its platforms, Foundry and Gotham. Deltas, the Forward Deployed Software Engineers, "deploy our software platforms to customers". The name Delta goes back to the company's early days, when each Business Development team was named after a letter of the NATO alphabet. The post places Deltas in Business Development and Devs in Product Development.

The same post put the difference in one line: a Dev's focus is "one capability, many customers", a Delta's is "one customer, many capabilities". It remains the sharpest definition of the job. Next to the Delta works the Deployment Strategist, known internally as "Echo", a role Palantir's blog places, in theory, closer to product management. Both tags still label postings on Palantir's job board in October 2026.

The acronym is in the filing for Palantir's 2020 stock market listing. Its S-1, filed with the SEC on 25 August 2020, says: "Our forward deployed engineers ('FDEs') have travelled to bases in Afghanistan and factories in the industrial Midwest to deploy our platforms." The same filing calls FDEs the company's "first line in identifying research and development opportunities". Palantir's current posting claims the role outright: "We pioneered this unique position."

What a forward deployed engineer does all week

The week runs on one loop: scope the problem, build inside the customer's systems, take it live, measure it and send what you learned back to the product team. OpenAI's San Francisco posting fits most of the loop into one sentence: "You will own discovery, technical scoping, system design, build, and production rollout, partnering directly with customer engineering and domain teams."

Scoping comes first and never really stops. A Palantir Delta wrote in 2021 that Deltas "engage in initial customer discussions to explore, identify, and scope out the customer's business problems". Ramp's FDE team turned it into a mantra: "always be scoping". The reason is practical. Colin Jarvis, then OpenAI's head of FDE, told The Pragmatic Engineer newsletter in August 2025 that what the customer describes in scoping often "doesn't match the data/system reality on the ground".

Then you build where the customer works. Anthropic's London posting asks its FDEs to "work within customer systems to build production applications with Claude models" and to deliver "MCP servers, sub-agents, and agent skills that will be used in production workflows". Google Cloud calls its Applied AI FDE the "Agent Engineer" and asks for evaluation pipelines and observability for agentic workloads.

At the AI labs this build has a toolkit of its own: agents, tool connections, model choice and evaluation runs. I wrote a separate guide to the AI version of the job, from agent design to evals.

The job runs past the demo. A Palantir Delta described being "there with the users when the applications go live". OpenAI measures its FDEs "through production adoption, measurable workflow impact, and eval-driven feedback that changes product and model roadmaps". Six of the eight postings read in full for this guide ask FDEs to feed what they learn in the field back to product and engineering.

Expect to travel. Palantir's New York posting says up to 25% of the time, Salesforce about 25%, Cohere 20 to 40%, Anthropic in London 25 to 50% and OpenAI up to 50%. Of those eight postings, only Ramp's names no travel figure.

Forward deployed engineer vs software engineer, solutions engineer and consultant

Four roles get mixed up in job searches. Here they are side by side, in the employers' own terms.

A product software engineer and an FDE share a skill set. They differ in how many customers they serve and what they answer for. Palantir lets people move between the two roles. Its 2019 post notes "a lot of movement between them", and the hiring manager who wrote it started as an FDSE. Deltas also write product code when a missing feature blocks a customer, and they coordinate it with the product team.

The solutions architect works before the deal, the FDE after it. Anthropic splits the two in its own postings. Its Applied AI Architect is "a Pre-Sales architect focused on becoming a trusted technical advisor" who travels occasionally. Its FDE works with the Post-Sales, Product and Engineering teams and travels 25 to 50% of the time. Commercially the FDE still sits close to sales. Salesforce files its FDE posting under the job category "Sales", and Google places its FDEs in the Google Cloud Go-To-Market team.

Palantir fights the consultant comparison hardest. "A common misconception is that a Delta is a consultant," its 2019 post says, and a São Paulo Delta adds that a Delta "does significantly more engineering and technical work than a consultant". Joe Schmidt of a16z sees it the other way. In a 2025 opinion piece he wrote that at AI application companies the job of setting up AI agents for customers often falls to "a professional services employee, sometimes rebranded as a forward deployed engineer". The newest moves lean his way. OpenAI launched its Deployment Company together with an agreement to acquire Tomoro, "an applied AI consulting and engineering firm". Salesforce requires prior customer-facing delivery experience in consulting, professional services or forward deployed engineering.

My line is simple. An FDE writes production code inside the customer's systems and answers for the result after go-live. A consultant who does both is doing FDE work. If you come from consulting, the same work reaches you through three arrangements: a job at a firm, a subcontract under an integrator or a direct client. Pick the arrangement before you pick the employer.

Forward deployed engineer salary: what the postings print

These are the ranges employers print in their own FDE postings, read on 2 October 2026. Each is an annual salary range exactly as printed. Several postings add equity as a word, never as a number, so total compensation stays unknown.

Location moves the range too. Salesforce prints $88,970 to $287,910 as its typical US range and the higher band in the table for California, New York and select metro areas.

The highest US dollar figure on any FDE posting read for this guide is $445,000 base, the top of the range for OpenAI's Forward Deployed Security Engineer in Washington DC. Headlines that put FDE pay at a million dollars a year have no posting behind them.

Experience sets the entry point. The postings ask for 1+ years after college at Palantir, 3+ preferred at Ramp, 4+ at Anthropic, 5+ at OpenAI and Scale AI, and 5 years at Google. Palantir also takes FDSEs straight from university, through new-grad roles and internships.

If you write software on a product team today, the gap to close is customer-facing delivery, and I mapped the path from a product engineering job into forward deployed work in a separate guide.

Forward deployed engineer jobs: who hires in 2026

This is a snapshot of the employers' own job boards on 2 October 2026. A posting count measures open listings on one day. It says nothing about how many FDEs a company employs.

• OpenAI. 24 postings with "Forward Deployed" in the title on its own board, in San Francisco, New York, Seattle, Washington DC, London, Madrid, Tokyo, Seoul, Singapore, Sydney and Abu Dhabi, with healthcare, legal and financial services variants.
• OpenAI Deployment Company. An OpenAI subsidiary launched on 11 May 2026 with "more than $4 billion of initial investment" to embed FDEs in client organisations. OpenAI agreed to acquire Tomoro and its roughly 150 FDEs and Deployment Specialists. Its own board lists six FDE postings, in New York, London, Paris, Singapore and Sydney plus a US cybersecurity role, and invites established teams of two to five to apply together.
• Anthropic. FDEs sit in the Applied AI team. Individual FDE postings were open in London, Paris and Munich, and FDE manager roles in London and New York. The London posting calls its hires "founding FDEs".
• Google Cloud. Its Applied AI FDEs join the Google Cloud Go-To-Market team. A search of Google Careers returned 16 US FDE postings.
• Salesforce. 66 FDE postings on its careers site. Salesforce runs FDEs in pods of "two FDEs and one Deployment Strategist", the Palantir pairing under the same name.
• Cohere. 15 postings with "Forward Deployed Engineer" in the title, including agentic platform, sovereign AI and public sector variants.
• Mistral AI. Forward Deployed Machine Learning Engineer and Forward Deployed AI Engineer roles in Paris, Montréal, Palo Alto, New York and Abu Dhabi, asking for fine-tuning, advanced RAG or agentic work.
• Scale AI. Forward Deployed Software Engineers for the US public sector, with an active TS/SCI clearance or eligibility for one.
• Palantir. 54 FDSE postings tagged Delta and 34 Deployment Strategist postings tagged Echo.

Search every title variant. Ramp hires a "Software Engineer, Forward Deployed". Palantir and Scale AI hire a "Forward Deployed Software Engineer". Google numbers its levels: FDE III, FDE IV. In India, OpenAI and Anthropic hire under Applied AI titles. None of their FDE-titled postings was located there on 2 October, and OpenAI's Applied AI Engineer posting for Delhi, Mumbai and Bangalore lists forward-deployed engineering among the relevant backgrounds.

On 2 October 2026 Anthropic announced the Claude Frontier Academy, which uses the same acronym for "Frontier Deployed Engineer", takes participants by nomination through their organisations and is a different thing from the forward deployed job.

Frequently asked questions