Three words,
two Indias

Palantir coined “Forward Deployed Engineer” for engineers stationed inside customer organisations, building against live problems in the customer’s own systems. Until 2016 the company employed more of them, known internally as Deltas, than software engineers.

OpenAI stood up a dedicated forward deployed engineering function in early 2025, the first frontier lab to organise the work as a team of its own.

On 4 June 2025 a16z partner Joe Schmidt called FDE the hottest job in startups. Twenty-two of OpenAI’s 311 open roles carried the title at the time.

Anthropic opened a Bengaluru office on 16 February 2026 under Irina Ghose, formerly managing director of Microsoft India, and began hiring applied AI architects and engineers. It described India as the second-largest market for Claude.ai.

On 30 June 2026 AWS committed $1bn to a forward deployed engineering organisation that sends pods of five or six engineers into customers for 45-day cycles. Two days later Microsoft launched Frontier Company with $2.5bn and about 6,000 experts.

In July 2026 TCS chief executive K. Krithivasan said the company wants 1 to 1.5% of its 593,798 associates working as FDEs, as many as 8,900 people. Two weeks later Tredence pledged 200. The three words had arrived in Indian services.

In April 2025 there were 643 open roles in the United States titled Forward Deployed Engineer.

Twelve months later the count was 5,330, a rise of 729%, with advertised pay from $170,000 to more than $200,000.

India’s live pool of strict-title roles is 250 to 270 at any moment. A broad match on the title reached about 585 open postings in July 2026; a tight match in September returned 93.

Indian demand rose roughly 800% across the first three quarters of 2025. The growth is real; the base is small.

This is the open market. Each dot stands for ten roles. The solid dots are the live pool of 270; the hollow ones extend to the broad-match figure of 585.

TCS has pledged that 1 to 1.5% of its associates will work as FDEs: between 5,900 and 8,900 people, drawn from a workforce of 593,798 at the end of June 2026.

Infosys is targeting about 6,000. At its June 2026 AGM it reported roughly 600 AI agents built and 4,800 AI projects under way, with about 90% of clients described as AI-aware.

The mid-tier follows. Tredence has committed to 200 domain-native FDEs over 12 to 18 months, Coforge is moving from about 20 to 100, and EPAM from about 50 to 100.

Every dot on the right is a target. No company has said how many will be new hires and how many retrained staff. The cluster on the left is what a candidate can apply to today.

Grouped by the kind of company, the market splits five ways. The opacity of each tile follows the confidence of the evidence behind it.

Two frontier labs hire in Bengaluru. OpenAI has posted a remote AI Deployment Engineer role for startups and runs a dedicated FDE organisation globally; Anthropic is hiring applied AI architects and engineers for its new office.

Platform vendors carry the most live strict-title postings. Databricks lists FDE and Senior FDE roles in Bengaluru, Google Cloud hires GenAI FDEs at levels III and IV, and Palantir, Salesforce, Atlassian, Turing and HackerRank all use the title.

Among Indian AI startups, Sarvam AI stands out with several FDSE tracks and clients including Tata Capital, SBI Life, CRED, IDFC and LIC. PromptQL, Avaamo, Meesho and Gnani.ai post the title too.

Services firms hold the pledges. TCS, Infosys, Wipro, Coforge, EPAM and Tredence have announced FDE headcount or practices; Accenture, Fractal and Cognizant carry FDE-titled requisitions.

Consulting rounds it out: McKinsey QuantumBlack hires forward deployment engineers up to principal level in Bengaluru, and BCG X in Mumbai and Gurgaon. Krutrim, Yellow.ai, Quantiphi and Ema are named in secondary lists, and no primary FDE posting was found for any of them.

Bengaluru holds about 45% of India’s FDE job descriptions. Sarvam, Databricks, Google Cloud, Anthropic, PromptQL, Avaamo, Gnani.ai, Meesho, HackerRank, Atlassian and McKinsey QuantumBlack hire here.

Hyderabad takes about 22%, led by Google Cloud, EPAM and the global capability centres.

Gurugram and the wider NCR account for about 15%, with BCG X, Coforge in Noida, the GCCs and Sarvam’s Delhi team on public-sector work.

Mumbai and Pune together take about 12%. BCG X sits in Mumbai; TCS and EPAM hire in Pune.

Chennai holds about 6%. Taken together, close to 70% of the country’s roles sit inside a Bengaluru–Hyderabad corridor. The shares are best read as directional.

The core demand is for two to five years of backend, data or cloud experience followed by a lateral move into customer-facing delivery. Sarvam’s FDSE band and Meesho’s FDE-II both sit here.

The ladder runs to principal. McKinsey QuantumBlack asks for 10 to 12 years; Databricks and Sarvam post senior tracks at six to ten. Entry-level roles exist at Wipro and Composio and are the exception.

Across 1,000 FDE postings AIM analysed, customer discovery appears in 55% and building AI systems in 37%; Python is named in 66% of postings and TypeScript in 35%.

Indian postings add RAG, agent frameworks, MCP servers, vector databases, evals, Kubernetes and enterprise systems from SAP to ServiceNow. Travel is written into the job: 40 to 50% at Avaamo, up to 50% at Commure, about 25% historically at Palantir.

Services practices anchor the bottom of the table: ₹10 to 18 lakh at entry, ₹18 to 28 lakh at three to six years, ₹28 to 45 lakh beyond that.

Indian AI startups pay ₹28 to 45 lakh mid-career, with ESOPs on top. Live postings sit higher: PromptQL advertises ₹45 to 67 lakh for a forward deployed analyst in Bengaluru, and Cartesia ₹70 to 90 lakh for a founding FDE.

Global labs hiring remotely into India top the table at ₹90 lakh to ₹1.5 crore for senior engineers. Market-wide averages sit far below, at ₹13.2 lakh across India and ₹17 lakh in Bengaluru, because most of the volume is in services.

Set against a Palantir FDSE in the US at $171,000 to $295,000, or a frontier-lab senior at $400,000 and above, India’s FDE premium over a comparable software engineer is 20 to 40%. It is uneven: some FDEs earn less than peers in pure engineering roles.

IIT Delhi launched an Advanced Certificate in AI Forward Deployed Engineering: six months online, ₹1,65,000 plus GST, with a curriculum split 70/30 between building and delivering. IIT Roorkee runs a PG certificate with Futurense, and IIT Gandhinagar retooled a residential AI curriculum around the role, with a cohort starting this month.

Private providers moved faster and cheaper. Krish Naik Academy’s bootcamp costs about ₹10,000 over five to six months; School of Core AI charges ₹65,000 for five; Supervity pays a stipend for a 12-week trainee programme; Scaler and fde.academy run their own tracks.

From ₹10,000 to ₹1,65,000 for a similar promise: a job with the three words in the title.

Most of these programmes launched within the past year. Placement outcomes, once published, will show whether the supply side is real.