Forward deployed engineer vs solutions engineer: how the roles differ

What is the difference between a forward deployed engineer and a solutions engineer?

The cleanest single-company pair we found is Cresta, a conversational AI vendor with both roles open ^[3][1]. Its Solutions Engineer, AI Agent posting describes a specialist who joins sales opportunities early, before a proof of value, to run workshops, lead use-case and API discovery, own the most complex demos and help build the return-on-investment case for buyers. It also names a handoff from pre-sale to deployment and asks the engineer to work on it with Cresta's forward deployed product managers and engineers ^[1].

Its Senior Forward Deployed Engineer posting reads as the deployment side of that handoff. The engineer develops, configures and deploys AI agents, builds integrations with external systems such as APIs, databases and CRMs, tunes performance in complex enterprise environments and plans milestones with internal teams. The posting asks for proficiency in Python and Golang ^[3].

For the definition and history of the forward deployed role itself, see what a forward deployed engineer is.

How do the two roles compare side by side?

The table summarizes the postings. It describes patterns in what these employers wrote, not a rule every employer follows.

Does the work start before or after the sale?

In these postings the solutions engineer comes first. Cloudflare's Senior Solutions Engineer posting has the engineer work with the sales organization from discovery and solution design through proof of value and technical validation. The engineer owns the pre-sales evaluation through technical close, then coordinates a clean handoff to customer success, professional services, support or implementation teams ^[2]. OpenAI's Ads Solutions Engineer posting calls the role the technical expert in the sales process, working with client partners to unlock new business and speed deal closure, though it also extends into onboarding and launch readiness for advertisers ^[8][8] OpenAI, “Solutions Engineer, Ads Solutions - San Francisco”, 21 Sep 2026..

The forward deployed side of Cloudflare is a different shape. Its Forward Deployed Engineer is embedded in the team of one strategic customer as the primary technical resource on multi-stage migrations, and stays with the account after the migration as an adviser ^[5][5] Cloudflare, “Forward Deployed Engineer (FDE)”, 5 Feb 2026.. OpenAI's version owns discovery, technical scoping, system design, build and production rollout with customer engineering teams ^[4].

The line is not absolute. Cresta's forward deployed engineer is also asked to run interactive demos and present proofs of concept to prospective customers ^[3]. "After the sale" is where the role's weight sits, not a rule.

Who owns production, and how much of the job is code?

In these postings, ownership of the running system is the clearest separator. OpenAI's posting says success is measured through production adoption and measurable workflow impact, and its forward deployed engineer owns technical delivery across multiple deployments from first prototype to stable production ^[4]. The Cresta and Cloudflare solutions engineer postings stop earlier and hand over to someone else ^[1][2].

Code follows from that. OpenAI asks for production-grade code written and reviewed across frontend and backend in Python, JavaScript or comparable stacks, alongside scoping and sequencing delivery, and one bullet says to contribute directly in the code when progress or clarity depends on it ^[4]. Cresta asks for proficiency in Python and Golang, and after a resume review its process begins with a self-paced coding assessment ^[3].

Solutions engineers are technical too, but the code serves the sale. Cresta lists Python for prototyping, scripting or demo customization as a plus ^[1]. Cloudflare prefers scripting, APIs or automation that make demonstrations and evaluations repeatable ^[2]. OpenAI's Ads Solutions Engineer posting asks for SQL, API use and debugging of implementation issues ^[8].

Do the roles carry sales targets, and how is pay structured?

No posting we read uses the word quota. The difference shows up in compensation. The three forward deployed postings list a salary range and equity, plus an annual bonus at Cresta, and mention no sales compensation plan ^[3][5][4]. Cresta's solutions engineer role is eligible for variable target compensation, and Cloudflare's says its range includes on-target incentive pay under its Sales Compensation Plan ^[1][2]. OpenAI's Ads Solutions Engineer posting lists salary and equity without mentioning commission ^[8].

Cresta's pair shows why these numbers cannot be ranked. The forward deployed figure is base pay before a bonus. The solutions engineer figure is a target total that already includes variable pay, so its base is lower than the headline suggests. Cloudflare's solutions engineer posting shows the split for one group of remote locations: a base of $163,240 to $224,840 plus a variable target of $48,760 to $67,160 ^[2]. Seniority is not matched either: Cresta's solutions engineer text describes a principal-level role and its forward deployed posting a senior one ^[1][3]. Location and specialty differ across the rows, so we do not average them or say which role pays more.

Official data cover only part of this. A search of O*NET OnLine on 23 October 2026 returned no occupation titled forward deployed engineer, so we found no government pay figure for the role under that title. For sales engineers, O*NET OnLine shows a 2025 median wage of $124,900 a year, or $60.05 an hour, from BLS data ^[6][6] O*NET OnLine, U.S. Department of Labor, “41-9031.00 - Sales Engineers”.. The BLS survey behind those figures, Occupational Employment and Wage Statistics, counts base rates, commissions and incentive pay, and leaves out stock bonuses, non-production bonuses and the employer cost of benefits ^[9][9] U.S. Bureau of Labor Statistics, Occupational Employment and Wage Statistics, “Frequently Asked Questions”.. The occupation is broad, with sample tasks that include selling material handling equipment and numerical-control machinery as well as computer systems ^[6], so treat it as a baseline for the category, not a benchmark for the postings above.

Where do the two titles blur?

Four places showed up.

• Pre-sale work inside forward deployed postings. Cresta's forward deployed engineer runs demos and proofs of concept for prospects ^[3].
• The code expectation varies under one title. Cloudflare's forward deployed posting makes Terraform proficiency a must-have and treats wider scripting as nice to have, and its day-to-day is architecture, configuration and advising one customer ^[5]. OpenAI's wants production-grade code across frontend and backend ^[4].
• Titles drift. The posting we read for OpenAI's role that spans pre-sales discovery through adoption is titled Applied AI Architect, while the Solutions Engineer posting we read there is for the ads product ^[7][8][7] OpenAI, “Applied AI Architect”, 18 Aug 2025.. A search for one title can miss the same work under another.
• One employer builds a feedback loop between them. Cresta asks its solutions engineer to bring what forward deployed colleagues learn about agent performance and edge cases back into pre-sale conversations ^[1].

Whether the forward deployed title is a real distinction or a relabel depends on the employer. Our reading is that where an employer posts both roles, as these three do, they describe different jobs, though the boundary is not clean. We did not read postings from employers that hire only one of the two, so this says nothing about them. That is an interpretation of seven postings, not a measured finding.

Where does an architect title fit?

Architect titles are a third label to watch. OpenAI's Applied AI Architect posting describes the senior technical owner for a named portfolio of customers, guiding them from pre-sales discovery and evaluation through deployment and adoption. An account director owns commercial strategy, and the posting says delivery teams, not the architect, own implementation once engaged. Its listing offers commission and equity, and it names solutions architecture, solutions engineering and AI deployment among the backgrounds it values ^[7].

In that case the architect follows the customer further down the timeline than the Cresta or Cloudflare solutions engineer postings do, but the posting leaves implementation execution to delivery teams. We read only one architect posting, so we cannot say how other employers use the title; apply the same tests below instead of trusting the label.

How can you tell which role a posting really describes?

These five questions are our own reading of the postings, not an industry standard.

1. Who is accountable when it breaks in production? The person, or a team they hand it to?
2. Does pay include a sales plan or on-target earnings? That points to the solutions side ^[1][2].
3. What does the process test first? A coding assessment, as in Cresta's forward deployed process ^[3], points to the build side.
4. How many customers? One embedded customer or a few deployments suggests forward deployed work; a stream of opportunities suggests solutions work.
5. Where does the work end? At a technical close and handoff, or at adoption and a stable production system?

What should engineers and recruiters take from this?

For an engineer who wants to own a system and its results, the OpenAI and Cresta forward deployed postings describe that job and ask for strong coding skills ^[4][3]. For someone who prefers discovery, demos and working with sellers, the solutions postings describe that and ask for five or more years of customer-facing technical work ^[1][2]. For a recruiter, the two postings ask for different backgrounds: a candidate with only pre-sales experience may not meet a posting that asks for production-grade code ^[4].

To weigh the forward deployed role against building AI products inside one company, see forward deployed engineer vs AI engineer.

Common questions