A forward deployed engineer (FDE) is a software engineer who works directly with a customer to get a product running inside that customer's real systems, then sends what they learn back to the product team. The job began at Palantir and spread to AI companies in 2026. Entry-level FDE roles in India are still rare.
A forward deployed engineer works inside the customer's world instead of only inside the product team. Searches for the role have jumped in India during 2026, and most explainers repeat each other. This page uses the employers' own words: public job boards and career pages read on 11 November 2026, plus Palantir's own explainer, which is where the title comes from.
Job counts come from each company's public board on that date: Palantir (304 openings), OpenAI (828) and Anthropic (677). Pay figures come from postings and Levels.fyi and are labelled. For longer reads, see the Masai blog guides Forward deployed engineer jobs in India, demand and who is hiring and What is a Forward Deployed Engineer? FDE vs Software Engineer vs Solutions Engineer.
Forward deployed engineer meaning, and what a forward deployed AI engineer is
Palantir created the job. In a 2019 post it splits its engineers into Devs and Deltas. A Dev builds a platform component that many customers use. A Delta, formally a Forward Deployed Software Engineer, deploys and customises the platform for one customer and is judged by that customer's outcome. Palantir sums it up as one capability for many customers versus one customer with many capabilities.
The post says the name Delta is a throwback to the early days, when business teams were named after NATO alphabet letters. It also says a Delta is not a consultant: a consultant tends to hand over a one-time analysis, while a Delta builds a long-term solution with the customer and sometimes contributes code back to the core product. The Pragmatic Engineer newsletter reported in August 2025 that until 2016 Palantir had more of these engineers than software engineers.
AI labs copied the model. OpenAI says its Forward Deployed Engineering team partners with customers to turn research breakthroughs into production systems. Anthropic's posting says FDEs embed with strategic customers to ship AI applications built on its models. Palantir also lists a Forward Deployed AI Engineer role, which is the same job with a focus on generative AI: owning strategy and implementation with customers.
What a forward deployed engineer does day to day
Four real postings read on 11 November 2026 show the pattern.
| Posting | What you own | What they ask for | Travel and setting |
|---|---|---|---|
| OpenAI, FDE, San Francisco (posted 11 Sep 2026) | Discovery, scoping, system design, build and production rollout with strategic customers. Success is production adoption and workflow impact | 5+ years of engineering or customer-facing technical work, Python or JavaScript, LLM systems built or deployed | Hybrid, up to 50% travel |
| Anthropic, FDE, London (updated 14 Sep 2026) | Production applications on Claude inside customer systems, plus MCP servers, sub-agents and agent skills for real workflows | 4+ years in a technical customer-facing role, production LLM and evaluation experience, Python | Travel 25 to 50% |
| Palantir, Forward Deployed AI Engineer, New York | End-to-end generative AI workflows taken to production, in small teams that work like a hands-on startup CTO | Production delivery with generative AI, data pipelines and analytics tools | Embedded in customer problems |
| Sarvam AI, Strategic Deployment Engineer, Delhi (posted 17 Apr 2026) | Full lifecycle of AI deployments, including air-gapped and on-premises sites, as the technical contact for each account | 3 to 6 years, Python, Docker, Linux, REST, CI/CD, LLM serving tools such as vLLM and Ollama, RAG | Embedded with clients |
The loop is the same in all four: learn the customer's real workflow, build something that works in their systems, get it adopted, and send what you learned back to the product team. Palantir's Deltas describe weeks that swing between writing code and scoping a project with a client.
FDE vs software engineer, solutions engineer, sales engineer and product manager
Titles overlap and companies use them differently. This table compares how the same employers describe each role in the postings above.
| Role | Focus in these postings | Judged by |
|---|---|---|
| Software engineer | One capability, many customers: owns a platform component end to end (Palantir's Dev) | Quality and reach of the product component |
| Forward deployed engineer | One customer, many capabilities: deploys and customises the product for a single customer, and writes code in their environment | Customer outcome, such as production adoption and workflow impact (OpenAI) |
| Solutions engineer or sales engineer | Technical expert in the sales process: discovery, integration advice and technical answers (OpenAI's Ads Solutions Engineer posting) | New business and deals closed |
| Applied AI Architect (pre-sales) | Technical advisor who guides enterprises from discovery to deployment. Anthropic's Bengaluru posting calls it pre-sales and asks for 10+ years | Customer adoption of the vendor's product |
| Product manager | Sets priorities and the roadmap for what many users get (OpenAI's API Agents PM posting) | Product outcomes across all users |
A practical test: if the job needs you to write and deploy code inside someone else's systems, it is an FDE job whatever the title says. If it ends when the customer signs, it is pre-sales. See also what an AI agent is, since many FDE projects now build agents.
Why AI companies hire FDEs in 2026
Fortune reported on 3 September 2026, citing Lightcast data first reported by The New York Times, that postings for forward-deployed engineers rose more than 1,000% between January and August 2026 against the same period a year earlier. Broader tech postings rose 13%. The article quotes the president of Dice saying these engineers help companies connect AI models to their own data, systems and workflows, which is where many AI projects stall.
The postings say the same in their own words. OpenAI asks FDEs to bring back eval-driven feedback that changes product and model roadmaps. Anthropic asks them to codify repeatable deployment patterns for its product and engineering teams. In India, Sarvam's Deployment Strategist posting says banks, insurers and government departments are not short of AI pilots but of AI in production.
Treat posting counts with care, because they depend on the title. Palantir's board had 99 openings with forward, deployed or deployment strategist in the title out of 304, including interns and new graduates. OpenAI's had 39 FDE-titled roles out of 828.
Forward deployed engineer jobs in India: what is actually open
As of 11 November 2026, neither OpenAI's nor Palantir's public board listed an FDE role in India. OpenAI listed FDE roles in Singapore, Tokyo, Seoul, Sydney and Abu Dhabi, among other cities. Anthropic listed FDE roles in Munich, Paris and London, and its Bengaluru and Mumbai roles are titled Applied AI Architect. Four India postings we opened:
| Employer and role | Place and date | Experience asked |
|---|---|---|
| Sarvam AI, Strategic Deployment Engineer | Delhi, posted 17 Apr 2026 | 3 to 6 years in software or ML engineering, with one full-cycle enterprise or on-premises deployment |
| Sarvam AI, Deployment Strategist | Bengaluru, posted 11 Aug 2026 | From 4+ years, with client-facing delivery you owned, often from consulting or systems integration |
| NTT DATA Services, FDE (AI and Agentic Systems) | Bangalore, dated 21 Oct 2026 | 8+ years in solution architecture, consulting, AI engineering or customer engineering |
| Anthropic, Applied AI Architect | Bangalore, updated 4 Nov 2026 | 10+ years in technical customer-facing roles |
None of the four is an entry-level role. Palantir's new-graduate forward deployed roles are in New York, London, Seoul, Chicago and its US government teams, not India. Based on what these postings ask for, a realistic route is two to three years of shipping software, a move into deployment or implementation work, then an FDE application.
Forward deployed engineer salary: sourced figures
| Source | Figure | Caveat |
|---|---|---|
| OpenAI posting, FDE, San Francisco (11 Sep 2026) | $185,000 to $300,000 a year plus equity | US pay for 5+ years of experience, not an India figure |
| Anthropic posting, FDE, London (14 Sep 2026) | £225,000 to £255,000 a year | UK pay, listed as annual salary |
| Lightcast data reported by Fortune (3 Sep 2026) | Median advertised US salary above $188,000, against about $145,000 for software engineers | Advertised, not paid, and US only |
| Levels.fyi, FDE, India (11 Nov 2026) | Median total compensation ₹17.6 lakh; 25th percentile ₹15 lakh; 75th ₹42.1 lakh; 90th ₹55.7 lakh | Self-reported, and the page does not show the sample size, so treat it as indicative |
The India postings above publish no pay range, and we found no AmbitionBox page for the FDE title. For the wider band, see AI engineer salary in India.
Skills, interview signals and a 90-day plan
Skills the postings ask for
- Production Python, plus one more language such as TypeScript or Java.
- LLM applications: prompting, agents, retrieval (RAG) and evaluation frameworks.
- Deployment basics: Docker, Linux, REST APIs, CI/CD, and serving tools such as vLLM or Ollama.
- Technical decomposition: breaking a business problem down to the code that solves it, as a Palantir Delta describes it.
- Judgement: deciding whether to build a feature into the product or ship something custom faster.
- Clear communication with engineers, product teams and customer stakeholders.
Interview signals
Sarvam's posting lists what it looks for: you have shipped and maintained a working system end to end where reliability mattered; you made a call on unclear client requirements without being told what to do; and you have open-source work, side products or founder experience that shows follow-through. Prepare two stories that prove this, with numbers.
A 90-day plan for a student or fresher
- Days 1 to 30: get comfortable with Python, Git, Docker and REST APIs. Build and deploy a Python API for free.
- Days 31 to 45: build a retrieval chatbot on messy real documents with the free Gemini and Streamlit guide, then write 20 test questions with expected answers. That is your first eval.
- Days 46 to 60: learn how agents call tools with the free Hugging Face AI Agents course, then connect an MCP server to your editor.
- Days 61 to 80: find one real user, such as a college office or a local shop. Run a discovery call, write down their workflow, build the smallest useful version and deploy it in their setting.
- Days 81 to 90: write a one-page case study covering the problem, what you built, what broke and the before and after numbers. Share the repo. More ideas are in agentic AI project ideas.
This plan builds the evidence of a first deployment. It does not make you eligible for the roles above, which ask for years of experience. For the wider path, start with the AI and ML learning path.
Frequently asked questions
What is a forward deployed engineer?
A forward deployed engineer (FDE) is a software engineer embedded with a customer to deploy and customise a product inside that customer's systems. The title comes from Palantir, where the role is called a Delta. AI companies such as OpenAI and Anthropic now hire FDEs to turn their models into production systems for large customers.
What is the FDE full form?
In tech jobs FDE stands for forward deployed engineer. Companies also write forward deployed software engineer (Palantir's FDSE), forward deployed AI engineer or deployment engineer. Read the posting: the work is deploying a product at a customer and building what is missing, not selling.
What is the difference between a forward deployed engineer and a software engineer?
A software engineer builds one platform component that many customers use. An FDE works with one customer and deploys, customises and sometimes extends the product for that customer's goal, then feeds what they learn back to the product team. Palantir describes it as one capability for many customers versus one customer with many capabilities.
Are there forward deployed engineer jobs in India?
A few. On 11 November 2026 we opened FDE-style postings at Sarvam AI (Delhi and Bengaluru) and NTT DATA Services (Bangalore), and an Applied AI Architect role at Anthropic in Bengaluru. They ask for 3 to 10 years of experience. OpenAI's and Palantir's public boards listed no FDE role in India.
What is the salary of a forward deployed engineer in India?
Levels.fyi shows a self-reported median total compensation of ₹17.6 lakh for India, with a 25th percentile of ₹15 lakh and a 75th of ₹42.1 lakh, as of 11 November 2026. The sample size is not shown. US and UK postings pay far more, but those are not India figures.
Can a fresher become a forward deployed engineer?
Directly, rarely in India. Palantir's new-graduate forward deployed roles are in New York, London, Seoul and Chicago, and Indian postings we read ask for three or more years. A fresher can build evidence first: one deployed project for a real user, with an eval and a case study.
What is a forward deployed AI engineer?
The same role with an AI focus. Palantir's Forward Deployed AI Engineer posting says the engineer owns generative AI strategy and implementation with customers, building end-to-end workflows and taking them to production.
Sources checked
- Palantir 2019 post on Devs and Deltas
- OpenAI FDE posting, San Francisco, pay and duties
- Anthropic FDE posting, London
- Anthropic Applied AI Architect, Bangalore
- Palantir Forward Deployed AI Engineer posting
- Posting growth and advertised pay
- India median total compensation
- NTT DATA Bangalore FDE posting
- Sarvam Strategic Deployment Engineer, Delhi
- Sarvam Deployment Strategist, Bengaluru
- OpenAI Ads Solutions Engineer posting
- OpenAI API Agents product manager posting
- Palantir FDE history
Learn this with mentors
AI and machine learning programmes with admissions open: live classes, mentors and project work.

