Blog
/
Career

Companies hiring forward deployed engineers: the living list

20 min read
Night city highway curving toward a distant skyline.
Table of contents
Section oneSection two

Forward deployed engineering is the job title the AI boom invented a shortage for. Business Insider reported that Indeed postings for forward deployed engineers were 543 percent above January 2025 levels by April 2025, and by April 2026 sat 5,230 percent above that baseline, roughly 729 percent growth in a single year. Job-market analyst Bloomberry independently counted postings growing 1,165 percent year over year. Whatever the exact multiple, the direction is the same: companies concluded that AI sells itself in demos and fails by itself in production, and the fix is an engineer deployed at the customer.

This page is the living list of who is actually hiring, what the role looks like at each company, and what the public record says about pay. We refresh it against careers pages and reporting on a rolling basis, so listings reflect the pattern of each company's hiring rather than a snapshot of one week's requisitions. If you are still deciding whether the role is for you, start with what a forward deployed engineer actually is and how the job differs from standard software engineering, then come back for the map.

What the role pays right now

Public numbers cluster into two tiers. For the broad market, Indeed data puts pay at roughly $170,000 to over $200,000. At the AI labs and the top of the market, total compensation runs $350,000 to $550,000 for mid-to-senior roles, with Bloomberry putting the median base salary at $173,816. Posted bands at specific companies land in between; Salesforce's multi-level New York range, for example, spans roughly $158,000 to $384,000 across levels. Two patterns worth knowing: customer-facing engineering now prices at or above pure product engineering at several of these companies, and the spread between levels is wide because the role's leverage grows with trust.

What a thousand job posts reveal about the role

Perspective AI analyzed 1,000 FDE job posts and the composite is remarkably consistent across companies. Core engineering appears in 95 percent or more of posts: Python and/or TypeScript, SQL, and cloud. AI-specific skills appear in 80 percent or more: LLM application development and retrieval. And the fastest-rising requirement, now in 70 percent or more of posts, is explicitly customer-facing: requirements discovery and customer empathy, written into an engineering job description. That discovery cluster is the quiet redefinition of the profession hiding under one job title: the posting is an engineering job that now writes customer empathy into the requirements.

The top of the market pays accordingly. Perspective's bands put mid-level FDEs at $300,000 to $450,000 total compensation, senior at $450,000 to $550,000, and staff or principal at frontier labs at $600,000 to $1.2 million and above, with equity representing 55 to 70 percent of total comp at the top. Read those numbers next to the failure statistics driving them, and the market's logic is plain: if an engineer at the customer is the difference between the 95 percent of AI pilots that fail and the 5 percent that do not, that engineer is cheap at any of these prices.

1. Palantir: where the role was born

Palantir coined the forward deployed engineer role and has run the largest, longest-standing forward deployed organization in the industry, listing roles under titles like Forward Deployed Software Engineer. The Palantir version is the deep-end archetype: engineers embed with government and commercial customers, live inside the customer's problem for months, and build on Palantir's platforms until the deployment works in that environment. It is the strongest brand an FDE resume can carry, precisely because the company has treated deployment as the product for two decades.

What to expect: high ownership, real field time, customer environments that range from banks to battlefields, and an interview process famous for testing how you decompose messy real-world problems rather than textbook algorithms. Browse open roles at palantir.com/careers.

2. OpenAI: deploying frontier models into enterprises

OpenAI is named in Business Insider's reporting among the companies hiring forward deployed talent as enterprise demand for AI tooling grows. The shape of the job follows from the business: enterprises buy access to frontier models and then need those models wired into their data, their compliance constraints, and their workflows, and forward deployed engineers are the ones doing the wiring at the accounts that matter most.

What to expect: the fastest-moving product surface in the industry, customer problems that have never been solved before by anyone, and the pressure that comes with deployments that double as flagship case studies. Openings at openai.com/careers.

3. Anthropic: forward deployed work on Claude

Anthropic appears in the same Business Insider reporting as a company hiring forward deployed engineering talent. The work centers on getting Claude producing real value inside enterprise environments: building the integrations, evaluations, and guardrails a specific customer needs before an assistant becomes part of their operations.

What to expect: enterprise deployments where safety and reliability requirements are part of the engineering problem rather than an afterthought, and customers who chose the vendor specifically for that posture. Roles at anthropic.com/careers.

4. Salesforce: the biggest company to bet the go-to-market on FDEs

Salesforce built a formal forward deployed engineering program around Agentforce, and it is the clearest sign the role has gone mainstream. Public requisitions like Forward Deployed Engineer, Agentforce for Supply Chain and multi-level FDE postings describe engineers who embed with a single enterprise customer for engagements of roughly three months and personally write the code that ships custom agents into production. Exponent's tracker reports the program launched in April 2025 and grew rapidly, with twenty-plus live requisitions across multiple continents by mid-2026 and posted New York bands of roughly $158,000 to $384,000 across levels.

What to expect: enterprise scale, a structured program rather than a frontier free-for-all, and one of the few genuine early-career paths into the role, with levels that run from entry to principal. For someone who wants FDE work with a big-company support system, this is currently the standout on the list.

5. Google Cloud: field engineering at hyperscale

Google Cloud is named in Business Insider's reporting among the companies hiring forward-deployed engineering talent. The cloud version of the role sits close to the customer at the intersection of AI platforms and enterprise systems: making Google's AI stack land inside organizations whose environments are as heavy as environments get.

What to expect: the largest enterprise customers in the world, deep platform surface area, and the resources of a hyperscaler behind you. Search current openings at careers.google.com.

6. Stripe: deployment engineering where the money moves

Stripe rounds out Business Insider's named list. Payments is an unforgiving domain for customer-specific work: every large merchant's stack is different, correctness is measured in money, and the engineers who wire Stripe into those environments carry real production stakes at some of the most demanding customers in software.

What to expect: high correctness culture, customer problems where the integration is the revenue path, and the credibility that comes from shipping in a domain where bugs have invoices. Roles at stripe.com/jobs.

7. Rippling: forward deployed work in the compound startup

Rippling appears among the companies where Paraform reports placing forward deployed talent. The company's compound-startup strategy, many products on one employee-data graph, generates exactly the customer-specific integration surface FDEs exist for: every serious customer connects that graph to their own systems in their own way.

What to expect: startup pace with real enterprise customers, and integration work that touches payroll, IT, and finance systems where mistakes are visible. Openings at rippling.com/careers.

8. Decagon and the AI-agent startups

Decagon, an AI customer-support agent company, is also on Paraform's placement list, and it stands in for a whole cohort: AI-agent startups whose product only works once it is woven into each customer's stack. At these companies the forward deployed engineer is often the difference between a pilot and a rollout, and early FDEs get the kind of ownership that turns into founding-adjacent equity stories.

What to expect: small teams, direct customer exposure from week one, and the highest variance on the list in both risk and upside. Start at decagon.ai.

9. Databricks: deployment engineering for the lakehouse era

Databricks appears on Perspective AI's list of companies hiring FDEs, and the work is what you would guess from the product: getting data and GenAI workloads live inside enterprises whose data estates resist generalization. Roles sometimes ship under adjacent titles like AI engineer or customer-facing engineer, so search the skill set, not just the phrase.

What to expect: heavyweight data problems, technical customers, and deployments where the integration surface is the customer's entire data platform. Openings at databricks.com/company/careers.

10. Snowflake: the same motion, data-platform flavored

Snowflake also appears on Perspective's hiring list, with solutions-oriented forward deployed roles around platform and AI deployment. The center of gravity is enterprises standardizing data and AI workloads on Snowflake and needing engineers who make that real account by account.

What to expect: enterprise data gravity, long-lived customer relationships, and integration work at the warehouse boundary. Roles at careers.snowflake.com.

11. Cohere: enterprise agents, deployed hands-on

Cohere, named on the same list, hires forward deployed engineers around its enterprise agent platform: building and shipping custom agents inside large organizations, with the security-conscious posture that defines Cohere's enterprise pitch.

What to expect: enterprise AI deployments with data-privacy constraints as a feature of the work, in accounts where the deployment is the product experience. Openings at cohere.com/careers.

12. Scale AI: forward deployed at the data frontier

Scale AI hires forward deployed engineers and data scientists, per Perspective's analysis, with role language famous for emphasizing ambiguity and first-principles problem solving for AI labs and enterprises.

What to expect: customer problems at the rawest end of AI deployment, often for the most sophisticated AI buyers in the market. Roles at scale.com/careers.

13. The vertical-AI cohort: Sierra, Harvey, Glean, Cresta, Hebbia, Writer

The fastest-growing slice of FDE hiring is the vertical AI application companies. Perspective's job-post analysis names Sierra, Harvey, Glean, Cresta, Hebbia, and Writer among them, alongside Decagon above. The pattern is structural: each sells an AI product whose value only materializes once it is woven into a specific customer's systems and workflows, so each hires engineers whose whole job is that weaving. At Glean the role centers on high-stakes onboarding where standard APIs run out; at Harvey it is legal environments; at Sierra and Cresta, customer-experience stacks; at Hebbia and Writer, knowledge-heavy enterprise workflows.

What to expect across the cohort: small teams, founding-adjacent ownership, direct exposure to whether the company's product actually lands, and the widest equity variance on this list. These are also the companies where one strong FDE visibly moves the revenue line, which is either pressure or leverage depending on your temperament.

14. The consultancies: when Deloitte posts the title, the title has arrived

The surest sign a role has crossed from frontier to mainstream is the consulting industry posting it verbatim. Deloitte has run public requisitions for a Forward Deployed Engineer focused on Databricks work, which tells you two things at once: enterprises are asking the big firms for forward deployed delivery by name, and the title now carries meaning in procurement conversations, not just in startup job boards.

What to expect: client-services rhythm rather than product rhythm, breadth across accounts rather than depth in one, and a strong on-ramp if you want the customer-facing craft before choosing a product company. The tradeoffs between consultancy-delivered and in-house forward deployed work are a big enough topic that we wrote a whole guide to them.

How to use this list

Fourteen entries is a map, not a strategy, so impose one. The companies above sort into three risk tiers, and the strongest applications run two per tier rather than fourteen scattershot.

Structured programs (Salesforce, Google Cloud, the consultancies): real leveling, onboarding, and early-career paths. Best if you are converting into the role from product engineering or graduating into it, and want the craft with a support system.

Established frontier (Palantir, OpenAI, Anthropic, Databricks, Snowflake, Stripe, Scale AI, Cohere): the role at full intensity inside companies whose deployments are existential to revenue. Best if you already ship end to end and want the strongest possible version of the work on your resume.

Vertical cohort (Sierra, Harvey, Glean, Cresta, Hebbia, Writer, Decagon, Rippling): maximum ownership and equity variance. Best if you want your individual work visible in the company's outcome, and can price the risk honestly.

Tailor per tier, not per company: the structured programs want evidence you can operate inside process, the frontier wants evidence you do not need any, and the cohort wants evidence you can be the process. Sequence loops so a favorite lands last; the customer-scenario round improves fast with live repetitions.

Where the role goes next

Three signals in the data say this list will be longer next year, not shorter. Salesforce built a leveled, bootcamped program around the title within a year of creating it. Deloitte is selling the role to clients by name, which means enterprise demand has reached procurement. And customer-facing discovery now appearing in 70 percent or more of FDE posts suggests the title is formalizing work that much of the profession already does without the name. The fastest-growing variant is the explicitly AI-flavored one, which we cover separately in the forward deployed AI engineer.

How to read an FDE job post

The title is not standardized, so search wide and read carefully. The same job ships as forward deployed engineer, forward deployed software engineer, forward deployed AI engineer, deployment engineer, and sometimes solutions engineer with a build mandate. Four things separate a real FDE role from a support role wearing the title:

  1. You write production code. The post says you will build, not configure. If the deliverable is documentation and enablement, it is a different job.

  2. You own a customer, not a queue. Real FDE work is account-shaped: one or a few customers, deeply, rather than tickets from everywhere.

  3. The customer's environment is the hard part. Look for language about integrating with customer systems, data, and security constraints. That is the actual craft.

  4. There is a path from field to product. The best programs feed what FDEs learn back into the roadmap, which is also what makes the role a career accelerant rather than a cul-de-sac.

One search note: the abbreviation is ambiguous outside tech contexts, so search the full phrase plus a company or platform name rather than the initials alone.

Preparing to land one

The good news about a role this new is that the playbook is still short. The job-post data above says exactly what to show: core engineering (Python or TypeScript, SQL, cloud), LLM application skills, and demonstrated customer-facing judgment, the three buckets appearing in 95, 80, and 70 percent of posts respectively. Interview loops are converging too: Salesforce's published process runs recruiter screen, hiring manager round on AI project experience, a technical round on agent design, and a customer-scenario panel, and Exponent's cross-company FDE interview guide documents similar shapes elsewhere. Companies consistently screen for the same trio: strong general engineering, comfort in ambiguous customer situations, and evidence you can ship end to end without a team around you. Our full preparation guide, how to become a forward deployed engineer, covers the skills, the portfolio, and the interview loops company by company.

One preparation edge worth naming: understand the operational side of the job before your first day, because your future team will be living it. The working reality of FDE teams is deploying, hosting, vaulting, versioning, and handing off customer-specific code, and the tooling for that job is young enough that we wrote the guide to it. Candidates who can talk concretely about how customer builds should be operated, not just written, read as seniors regardless of years. And if you land the seat and inherit the laptop-era version of the job, Archway is the platform for FDE teams: the first two bridges are free, which is exactly enough to make your first month look like someone's first year. If you are the person standing the function up, not just filling a seat, how to run an FDE team is the operating manual, and the FDE product page is the shortest path to trying the backbone on a real customer build.

How to spot a healthy program from outside

Two companies can post the identical FDE title and offer opposite jobs, so evaluate the program, not the posting. Five signals are checkable before or during a loop. Leveling exists: the company posts the role at multiple levels or names a career path, which means someone designed a future for the seat rather than a staffing patch. The loop includes a customer scenario: programs that interview for the actual craft run one; programs that only run coding rounds are hiring a product engineer they plan to surprise. Engineers in the role are reachable: healthy programs let candidates talk to a working FDE, and what that person says about deploy paths and on-call reality is worth more than the whole careers page. The operational questions have answers: ask where customer builds run and who holds credentials, and grade the answer against the five verbs; a program running on laptops in year two of the boom is telling you about its priorities. And the team feeds the product: ask for an example of field work that changed the roadmap, because the programs where that loop runs are the ones where FDE experience compounds into influence rather than tickets.

None of these are gotchas. They are the questions a serious candidate asks anyway, and the companies on this list that run strong programs answer them happily, which is itself the signal.

About this living list. Companies enter when hiring is verifiable through a careers page, a public requisition, or named press reporting, and each entry links its evidence so you can check the current state yourself before applying. Openings churn weekly; patterns hold for quarters. If a company on this list stops hiring the role, we say so in the entry rather than silently deleting it, because the history of who built FDE teams is itself useful signal about who takes deployment seriously.

Sources and notes

A long list of live bridges: approval rules, usage reports, data migrations, SLA alerts, reconciliation, provisioning and compliance exports

Deliver hyper-personalized software at scale

Start free

Product

ProductSecurityPricing

Legal

Terms of UsePrivacy PolicyPurchase Terms

© 2026 Archway. All rights reserved.