Chuangu Think Tank Researcher Huang Ailin
In the AI era, the FDE (Forward Deployed Engineer) role has taken off. Data from Maimai, a Chinese professional networking platform, shows that FDE-related job postings surged 21-fold year-over-year in the first half of 2026, with major companies like ByteDance and Tencent following suit in creating such positions. According to the latest "FDE Industry Research Report" (hereinafter "the report") released by Xiaguang Think Tank, the number of FDE job openings in the US skyrocketed from 643 in April 2025 to 5,330 in April 2026, up 729% year-over-year, far exceeding the roughly 45% growth rate for AI engineers overall. 565 companies worldwide are hiring FDEs, with a median base salary in the US of $193,000, and total compensation at top companies can exceed $500,000. The supply-demand ratio is about 1:2.7, with a talent gap of over 60%.
FDE stands for Forward Deployed Engineer. The report defines it as a hybrid role that goes deep into customers' business sites, converts AI model capabilities into runnable business outcomes, and feeds frontline experience back into product and platform capabilities. They not only write production code, build systems, and handle deployments, but also communicate with customer management and frontline staff, turning vague "we want to use AI" requests into workflows that actually run. Palantir (a Silicon Valley tech company) pioneered the FDE role in 2010, serving defense and government clients, and over more than a decade of practice validated the commercial viability of the "engineers on-site + product feedback" model. 2025 became a key inflection point for FDE development, as OpenAI and Anthropic, with investments in the billions of dollars, elevated FDE from a supporting function to a "strategic business unit."
In terms of the current landscape, Palantir leads with its "platform + services" model, generating over $500,000 in annual revenue per employee with a gross margin above 80%; OpenAI follows closely with its "model + delivery" model; while consulting giants like Accenture have entered the field through staff outsourcing. Looking at representative cases, the commercial loop has been proven: Palantir delivered the TITAN system to the US Army, OpenAI partnered with JPMorgan Chase to deploy financial AI, and Anthropic teamed up with Fujitsu to expand into the Japanese market. The Chinese market is currently driven mainly internally by major companies like ByteDance, Alibaba, and Tencent, with external professional services still in their early stages; it is expected that after 2027, as these companies productize their capabilities, they will accelerate their outward expansion.
The report predicts that FDE will evolve through three stages: in the short term (1–2 years), it will remain a scarce role with salaries continuing to rise; in the medium term (3–5 years), it will transition toward scaled delivery; in the long term (5+ years), it will become a standard enterprise capability, with delivery capability becoming the core competitiveness of AI companies. The industry is still grappling with four major controversies, including the sustainability of business models, the risk of AI replacing itself, and the tension between scale and quality.
Accordingly, the report recommends: companies should make delivery capability a core part of their AI strategy and build a hybrid "platform + services" model to boost customer stickiness; individuals should shore up gaps in business understanding and communication, transitioning toward hybrid talent profiles; investors should watch for delivery service providers with vertical industry know-how; and the Chinese market should seize the window for AI going global, using FDE as a supporting delivery layer to break through overseas localization bottlenecks—particularly in emerging markets like Southeast Asia, the Middle East, and Latin America, where structural opportunities exist.
【Unauthorized reproduction is strictly prohibited! Contact: 028-86968276】