Last updated : added the requirements index from 563 live FDE engineer postings, with a download
Employers ask Forward Deployed Engineers for production software skills, applied AI and agent work, evaluation, systems integration, customer discovery, and a clear travel or on-site plan. The requirements below are quoted from live postings checked on October 22, 2026.
Thirteen employer postings were read in full for this page. FDE Dispatch has no affiliation with these employers. Postings close; confirm each linked careers page before you treat a quote as current.
Requirements index: share of 563 live FDE postings that mention each skill
Python is the most-mentioned skill in Forward Deployed Engineer postings: 76.2% of 563 live engineer-title FDE postings from 115 employers name it, pulled October 25, 2026. Customer-facing work (74.1%) and APIs or integrations (69.8%) come next. Where a posting states years of experience, the median first stated figure is 5 years. Hiring managers writing a job description can start from this index on How to hire a Forward Deployed Engineer.
| Requirement | Postings (of 563) | Employers (of 115) | What counts as a mention |
|---|---|---|---|
| Python | 76.2% (429) | 73.9% (85) | Python |
| Customer-facing work | 74.1% (417) | 73.0% (84) | customer-facing, client-facing, or working directly with customers or clients |
| APIs or integrations | 69.8% (393) | 80.9% (93) | API, APIs, integration, integrate with or into |
| Travel | 58.8% (331) | 45.2% (52) | travel, travels, traveling (travel stipends and reimbursement excluded) |
| TypeScript or JavaScript (incl. Node.js) | 56.5% (318) | 40.9% (47) | TypeScript, JavaScript, Node.js |
| LLMs or generative AI | 43.9% (247) | 62.6% (72) | LLM, large language model, generative AI, GenAI, foundation model |
| AI agents or agentic systems | 41.0% (231) | 61.7% (71) | agent, agents, agentic (recruiting, support and other human agents excluded) |
| Data pipelines or ETL | 36.1% (203) | 34.8% (40) | ETL, ELT, data pipeline, data engineering, data integration |
| AWS, GCP or Azure | 35.9% (202) | 37.4% (43) | AWS, Amazon Web Services, GCP, Google Cloud, Azure |
| Java | 24.0% (135) | 13.9% (16) | Java (JavaScript excluded) |
| Computer science degree | 22.2% (125) | 33.0% (38) | a degree word (degree, bachelor's, BS, master's, MS, PhD) next to computer science |
| RAG, retrieval or vector search | 19.4% (109) | 27.0% (31) | RAG, retrieval, vector database or vector search, embeddings |
| Kubernetes or Docker | 19.2% (108) | 38.3% (44) | Kubernetes, K8s, Docker |
| Evals (LLM or model evaluation) | 18.3% (103) | 33.0% (38) | evals, LLM or model evaluation, evaluation harness or framework (evaluate alone excluded) |
| Startup or founder experience | 11.9% (67) | 21.7% (25) | startup, early-stage, founder or founding engineer next to experience, background, prior or worked |
| SQL | 10.5% (59) | 20.9% (24) | SQL as a word (PostgreSQL, MySQL, NoSQL excluded) |
| Go (Golang) | 7.1% (40) | 14.8% (17) | Golang, or Go listed next to another language (go live, go-to-market excluded) |
- Each employer once. Databricks (87 postings) and Palantir (70) supply 157 of the 563 postings. The employers column counts each of the 115 employers once, so one employer's posting volume does not change it. APIs or integrations lead on that measure (80.9% of employers), followed by Python (73.9%) and customer-facing work (73.0%).
- AI work. 43.9% of postings mention LLMs or generative AI and 41.0% mention AI agents or agentic systems. By employer the shares are 62.6% and 61.7%. RAG or retrieval appears in 19.4% of postings and evals in 18.3%.
- Experience. 328 postings (58.3%) state a years-of-experience figure. The median first stated figure is 5 years, with the middle half between 4 and 6. Counting each of the 77 employers that state one once, the median is 5 years.
- Degrees. 22.2% of postings (125) mention a computer science degree.
Method and limits
FDE Dispatch took the 565 engineer-title postings in its October 24, 2026 snapshot, internships excluded, and re-fetched each posting's full text from its public Greenhouse, Lever, Ashby or Workday board on October 25, 2026. Two had closed, leaving 563. A posting counts for a row when its text matches that row's keywords at least once. About-the-company sections, equal-opportunity text and benefits lines are removed before matching, so a company tagline that mentions agents does not count. A mention is not a hard requirement: matches include nice-to-have lists and responsibilities. Keyword matching is approximate. Deployment strategist and FDE management titles are left out. The salary page's travel share (59.3% of 725 postings) includes those titles; on the 563 engineer-title postings here it is 58.8%. Employers that post only on LinkedIn, Indeed or other applicant systems are missing.
Download the per-posting flags: CSV, JSON and method and full keyword list. 563 rows, one per posting, each linked to its source. All published datasets are listed on FDE Dispatch job-posting data.
What changed: on October 25, 2026 FDE Dispatch added this requirements index and its download. The thirteen quoted postings below are unchanged from the October 22, 2026 check. Correction, October 25, 2026: degree abbreviations such as B.S. had hidden 3 computer science degree mentions; the degree row now reads 125 postings, up from 122.
Which postings mention which skill areas
Across the thirteen postings read for this page on October 22, 2026, every listing addresses production engineering and customer discovery; AI and agents, evals, integration, travel, and experience floors vary by employer. A cell marked yes means the posting text addresses that skill area; a dash means that area was not stated in the text we read.
Production software engineering
FDE postings treat you as an engineer who ships and operates production systems, with full-stack or multi-language coding and ownership from prototype to stable production.
- OpenAI, Forward Deployed Engineer (FDE) - SF: "Write and review production-grade code across frontend and backend using Python, JavaScript, or comparable stacks"
- Anthropic, Staff Forward Deployed Engineer: "Extensive experience building and operating production systems"
- Scale AI, Forward Deployed Software Engineer, Public Sector: "Strong production coding in Python and TypeScript (or comparable modern languages)"
- Salesforce, Forward Deployed Engineer: "Own technical components end-to-end: from architecture choice to deployed, validated, and observable in production"
- Databricks, Forward Deployed Engineer, FDE (L4): "delivering production-grade systems, designing and building reference architectures, custom applications and data ingestion and ML/AI model integration"
- Ramp, Software Engineer, Forward Deployed: "Track record of shipping high-quality products at scale"
- 84.51°, Lead Forward Deployed Engineer (FDE): "5+ years of experience producing high-quality software on Agile development teams"
Applied AI, LLMs, and agent work
Most of these postings name LLMs, agents, or generative systems as part of the build. How hard they require prior agent shipping varies by employer.
- OpenAI, Forward Deployed Engineer (FDE) - SF: "Have built or deployed systems powered by LLMs or generative models and understand how model behaviour affects product experience"
- Anthropic, Forward Deployed Engineer (London): "Production experience with LLMs including advanced prompt engineering, agent development, evaluation frameworks, and deployment at scale"
- Anthropic, Forward Deployed Engineer (London): "Deliver technical artifacts for customers like MCP servers, sub-agents, and agent skills that will be used in production workflows"
- Anthropic, Staff Forward Deployed Engineer: "You don't need to have shipped agents to production. You do need deep experience building and operating production systems, hands-on time with agents, and the judgment to lead architecture and advisory work with enterprise customers."
- Ramp, Software Engineer, Forward Deployed: "Extensive experience building with and for agents"
- Scale AI, Frontier Agents Engineer (Forward Deployed Engineering): "Design scalable agent architectures that combine LLMs, retrieval, memory, tools, structured knowledge, and enterprise data into reliable production workflows"
- Salesforce, Forward Deployed Engineer: "Design agent intelligence: prompts, reasoning, tool calls, and integration with customer systems, leveraging our Agentic AI platform and current LLM techniques"
Evaluation
Several postings name evals, quality metrics, red-teaming, or eval-driven feedback as part of the job. See evaluation suite for the FDE Dispatch definition.
- OpenAI, Forward Deployed Engineer (FDE) - SF: "You will measure success through production adoption, measurable workflow impact, and eval-driven feedback that changes product and model roadmaps"
- OpenAI, Deployment Lead, Forward Deployed Engineering (FDE) - Platform: "Can reason through integrations, data, evaluations, and enterprise deployment constraints"
- Scale AI, Frontier Agents Engineer (Forward Deployed Engineering): "Deploy evaluation harnesses using offline benchmarks, online experiments, golden datasets, regression suites, and LLM-as-a-Judge to detect quality regressions before they impact customers"
- Ramp, Software Engineer, Forward Deployed AI Solutions: "Evaluation plan covering quality metrics, acceptance tests, and red-teaming"
- Salesforce, Forward Deployed Engineer: "You ship quality code and evaluate AI outputs with engineering rigor"
Map these skill areas to the interview rounds that test them.
Systems integration and data
FDE work lands inside the customer's stack: APIs, warehouses, auth, cloud accounts, and data pipelines. Postings spell that out as architecture and integration ownership.
- Scale AI, Frontier Agents Engineer (Forward Deployed Engineering): "Architect and deploy production AI systems that integrate seamlessly into complex enterprise environments, including cloud platforms, data warehouses, internal APIs, business applications, and proprietary software systems"
- Salesforce, Forward Deployed Engineer: "Build and integrate data pipelines on Salesforce Data 360, Snowflake, Databricks, and customer data platforms"
- Databricks, Forward Deployed Engineer, FDE (L4): "implement end-to-end systems spanning data engineering, AI, and application development"
- Ramp, Software Engineer, Forward Deployed AI Solutions: "Integration plan across Ramp and customer systems"
- Palantir, Forward Deployed Software Engineer: "Wrangling massive-scale data and using AI to accelerate and enhance critical operations"
- Scale AI, Forward Deployed Software Engineer, Public Sector: "Build robust backend systems and data integrations that connect customer environments to Scale's platform and back"
Customer discovery and communication
These roles sit with the customer. Postings name discovery, stakeholder work, and clear communication with technical and business audiences. See discovery and FDE vs. solutions engineer for how ownership differs from a pure solutions seat. For the software-engineer comparison, see Forward Deployed Engineer vs. software engineer.
- OpenAI, Forward Deployed Engineer (FDE) - SF: "You will own discovery, technical scoping, system design, build, and production rollout, partnering directly with customer engineering and domain teams"
- Anthropic, Staff Forward Deployed Engineer: "Lead discovery and advisory engagements independently, shaping scope and technical strategy before a build starts"
- Anthropic, Forward Deployed Engineer (London): "Strong communication skills to conduct discovery with customers and to convey technical concepts to diverse stakeholders"
- Ramp, Software Engineer, Forward Deployed AI Solutions: "Partner directly with customers to understand current workflows, constraints, systems, data quality, and adoption blockers"
- Salesforce, Forward Deployed Engineer: "You have prior customer-facing technical delivery experience (consulting, professional services, or forward deployed engineering) — this is a requirement, not a nice-to-have"
- Palantir, Forward Deployed Software Engineer: "Engaging directly with customer stakeholders, from technical teams to executives"
- Scale AI, Forward Deployed Software Engineer, Public Sector: "Embed with government customers to understand their mission, scope the problem, and turn it into working software"
Travel and on-site expectations
When a posting names travel, the stated range on these listings runs from 20% to about 50%. Internal embeds can be office-heavy without a travel percentage.
- OpenAI, Forward Deployed Engineer (FDE) - SF: "Travel up to 50% is required"
- Anthropic, Staff Forward Deployed Engineer: "Travel frequently (25%) to customer sites to build in person with customers"
- Anthropic, Forward Deployed Engineer (London): "Travel frequently (25-50%) to customer sites to build in person with customers"
- Ramp, Software Engineer, Forward Deployed AI Solutions: "Willingness to travel up to ~50% as needed, flexible based on project needs and client needs"
- Salesforce, Forward Deployed Engineer: "You're excited to travel ~25% of the time, working directly alongside customers"
- Databricks, Forward Deployed Engineer, FDE (L4): "Travel to customers 20% of the time"
- Scale AI, Forward Deployed Software Engineer, Public Sector: "Expect 25% travel with up to 50% travel as necessary, and/or relocation to Washington DC or St. Louis"
- Palantir, Forward Deployed Software Engineer: "Willingness and interest to travel as needed to client sites. Ability to travel 25-50% preferred, but requirements vary by team and location"
- 84.51°, Lead Forward Deployed Engineer (FDE): "84.51° follows a 5‑day in‑office work schedule to support collaboration, alignment, and team connection"
Experience level
Stated floors on these listings: 3+ years at Ramp (preferred) and Salesforce; 4+ at Anthropic London and Scale Frontier Agents; 5+ at OpenAI, Scale Public Sector, and 84.51°; 8+ on Anthropic Paris; 9+ on Databricks L4. Anthropic Staff and Palantir Seoul do not put a single numeric years floor in the same form. Years bands, early-career titles, and employers asking for 2 or fewer years are indexed on Entry-level Forward Deployed Engineer jobs. Title strings and seniority labels also vary; see Forward Deployed Engineer job titles.
- OpenAI, Forward Deployed Engineer (FDE) - SF: "Bring 5+ years of engineering or technical deployment experience that includes customer-facing work"
- Ramp, Software Engineer, Forward Deployed: "3+ years of software engineering experience preferred"
- Salesforce, Forward Deployed Engineer: "You have 3+ years of software engineering or technical delivery experience"
- Scale AI, Frontier Agents Engineer (Forward Deployed Engineering): "4+ years of software engineering experience with strong fundamentals in distributed systems, data structures, algorithms, and system design"
- Databricks, Forward Deployed Engineer, FDE (L4): "9+ years of experience in data engineering, data platforms & analytics, or software engineering"
How to build each skill area
Build each skill area with one measured deployment and the practice targets below, matched to the employer quotes on this page. The full path is laid out in How to become a Forward Deployed Engineer.
- Production software engineering. Ship one system someone uses for real work, with monitoring and an owner after launch. Keep a short write-up of what broke after go-live. Rehearse that story with the practice questions for deployment deep dives.
- Applied AI and agents. Build one agent or LLM workflow that calls tools against a real process, with a clear failure mode. Anthropic's Staff posting allows hands-on agent work without prior production agent shipping; other postings require production LLM experience. Match the posting you want.
- Evaluation. Write an evaluation suite: baseline cases, acceptance tests, and a measured before-and-after. Scale and Ramp name harnesses, golden sets, and red-teaming by name.
- Systems integration and data. Practice one integration into a customer-like stack: auth, an API or warehouse, and a data path you can defend in a customer-scenario system design round.
- Customer discovery and communication. Run discovery interviews, write a one-page scoped brief, and practice stakeholder pushback with the interview loop overview and field guide. Salesforce treats prior customer-facing delivery as required.
- Travel and on-site. Read the travel line on the posting before you apply. Ranges on this page run from 20% to about 50% when stated. Internal embeds can be office-first instead.
- Experience level. Compare your years and customer-facing history to the posting floor, then to the title variants on the job titles page. Close gaps with a measured deployment that shows the customer-facing and production work the posting names.
Related guides
- What is a Forward Deployed Engineer?
- How to become a Forward Deployed Engineer
- Entry-level Forward Deployed Engineer jobs
- The FDE interview loop, explained
- FDE interview loop field guide
- FDE interview practice questions
- Forward Deployed Engineer job titles
- Forward Deployed Engineer jobs by location
- Forward Deployed Engineer resume
- Forward Deployed Engineer salary
- How much do forward deployed engineers travel?
- How to hire a Forward Deployed Engineer
- How to evaluate a Forward Deployed Engineer training program
- FDE vs. solutions engineer
- Forward Deployed Engineer vs. software engineer
- Deployment strategist vs forward deployed engineer: what the paired business-side role does, who posts it, and how travel and pay compare
- FDE glossary: short definitions of Forward Deployed Engineer terms
Questions people ask
What skills do you need to be a Forward Deployed Engineer?
Across 563 live engineer-title FDE postings from 115 employers, pulled October 25, 2026, the most-mentioned requirements are Python (76.2% of postings), customer-facing work (74.1%), APIs or integrations (69.8%), travel (58.8%) and TypeScript or JavaScript (56.5%). Where a posting states years of experience, the median first stated figure is 5 years. The thirteen postings quoted on this page, checked October 22, 2026, group the requirements into seven areas: production software engineering, applied AI and agent work, evaluation, systems integration and data, customer discovery and communication, travel or on-site work, and a stated experience floor.
Do Forward Deployed Engineer roles require prior agent shipping experience?
It depends on the posting. Anthropic's Staff Forward Deployed Engineer posting states you do not need to have shipped agents to production, while still expecting hands-on time with agents and deep production-systems experience. Anthropic's London Forward Deployed Engineer posting asks for production experience with LLMs including agent development. Ramp's Core FDE posting asks for extensive experience building with and for agents. Read the linked posting for the role you want. Across 563 live engineer-title FDE postings pulled October 25, 2026, 41.0% mention AI agents or agentic systems; a mention is not a requirement to have shipped agents.
How much travel do Forward Deployed Engineer jobs require?
On the postings checked October 22, 2026 that name travel: Databricks lists 20%; Anthropic Staff lists 25%; Salesforce lists about 25%; Anthropic London and Anthropic Paris list 25-50%; Palantir Seoul FDSE prefers 25-50%; OpenAI FDE SF and Ramp Forward Deployed AI Solutions list up to about 50%; Scale Public Sector expects 25% with up to 50% as necessary. Some postings, including Ramp Core FDE and 84.51° Lead FDE, do not state a travel percentage in the text we quoted. Across 563 live engineer-title FDE postings pulled October 25, 2026, 331 (58.8%) mention travel.
Train against the skills employers post
FDE Dispatch maps what live FDE postings ask for, then trains engineers against the interview rounds those hiring loops use.