From Testing to Forward Deployed Engineering: A 2026 Fitment Guide for QA Engineers, SDETs & Analysts
Published September 4, 2026·10 min read
TL;DR
"QA" on a résumé hides three different jobs. A quality analyst checks software by hand and files defects (rarely codes). An SDET is a developer who builds the test automation. A QA engineer sits in the middle — runs and extends the SDET's tooling, writes some tests, still checks by hand. In Dexity's analysis of 1,243 US testing job descriptions, they pay very differently: analyst $73K–$96K, SDET $100K–$140K, QA engineer $110K–$149K. And nearly 6 in 10 testers we spoke with want out of testing entirely — which is where the AI-era opportunity comes in: forward-deployed-engineering JDs are dominated by the exact skills testing teaches (observability, testing, guardrails, evals, accuracy), at a median near $200K. Here's how the three roles differ, what they pay, and who can make the jump.
What's the difference between a QA analyst, QA engineer, and SDET?
They're three different jobs, and a job title won't tell you which one someone does. A quality analyst checks that software does what the requirements said — by hand, from a written test list — and logs defects; most of the day is in Jira, and they rarely write code. An SDET (software development engineer in test) is a developer whose product is the testing: they build the automation that drives the app and checks the answers on every code change. A QA engineer is the middle — and the largest group — owning quality for one part of the product, running and extending the SDET's tooling, writing simpler tests, and still checking plenty by hand. In Dexity's analysis of 1,243 US testing job descriptions, they pay $73K–$96K, $100K–$140K, and $110K–$149K respectively. The whole question is where on that chain someone sits.
Key facts
- "QA" covers three distinct roles — quality analyst, QA engineer, and SDET — separated by how much they build versus check.
- In Dexity's analysis of 1,243 US testing job descriptions: quality analyst $73K–$96K, SDET $100K–$140K, QA engineer $110K–$149K.
- Coding is the dividing line: a language is named in 23% of quality-analyst JDs, about half of QA-engineer JDs, and 76% of SDET JDs.
- In Dexity's conversations with 80 testers, 58.8% want to move out of testing entirely; another 42.5% want to stay but keep pace with AI.
- Across 2,110 forward-deployed-engineer JDs, the top asks are the testing skill set: observability 30.8%, testing 24.7%, guardrails 22.6%, evaluation 18.8%, accuracy/hallucination-catching 14.4%.
- Yet only 35 of those 2,110 FDE JDs mention "SDET" or "test automation" — the title doesn't exist, but the skill is wanted everywhere.
- The pay gap is large: forward-deployed roles run a median near $200K–$205K, versus $73K–$149K across testing.
QA analyst vs QA engineer vs SDET: the chain
The clean way to see it is as a chain. The SDET builds the test machinery, the QA engineer runs and extends it, and the analyst checks by hand and reports what's broken.
| Quality Analyst | QA Engineer | SDET | |
|---|---|---|---|
| Core job | Check by hand, file defects | Own quality for one area; run + extend tests | Build the test automation |
| Where they sit | End of the chain | The middle (largest group) | Start of the chain |
| Codes? | Rarely | About half do | Yes — it's a dev role |
| Typical tools | Jira, SQL, defect tracking | Selenium, Playwright, CI/CD, Python, SQL | Python, Java, TypeScript, Playwright, Jenkins |
Notice the tools column is the same chain again: the analyst's tools are for tracking work, the SDET's are programming languages. If you're trying to place yourself (or a candidate), three questions settle it: what do you work on day to day, how much of that is writing code, and are you building the test tooling or using it?
What does each testing role pay in 2026?
From Dexity's analysis of 1,243 US testing job descriptions (July 2026):
| Role | Pay range | Language named in JDs |
|---|---|---|
| Quality analyst | $73K–$96K | 23% |
| SDET | $100K–$140K | 76% |
| QA engineer | $110K–$149K | ~50% |
The interesting wrinkle: QA engineers top the range, slightly above SDETs, because they carry both ownership of a product area and enough automation skill — but the SDET's near-universal coding requirement (76%) is what makes it the most transferable role of the three.
What do testers actually want in 2026?
This is the part that reframes the whole field. In Dexity's conversations with 80 testers:
| What they want | Share |
|---|---|
| To move out of testing, into another domain | 58.8% |
| To stay relevant as AI takes over their work | 42.5% |
| Better pay | 26.2% |
| Job security (worried, or already laid off) | 10.0% |
Read the top two together, because they point in different directions. Nearly six in ten want out of testing altogether. About four in ten want to keep the job they have but not get left behind by AI. Both are rational responses to the same pressure — AI is automating the most manual, checklist-driven testing work first — but they lead to different next moves.
Why testing skills transfer to forward-deployed engineering
For the testers who want out, the AI era opened a door that fits them unusually well. A forward deployed engineer puts AI into a customer's business — and AI doesn't behave like normal software, because it can answer the same question two different ways. Someone has to decide whether it's good enough to switch on, and keep checking once it's live. That is, almost exactly, the testing mindset.
The FDE job descriptions bear it out. Across 2,110 forward-deployed-engineer JDs in Dexity's collection:
| What FDE JDs ask for | Share |
|---|---|
| Watching a live system and tracing problems (observability) | 30.8% |
| Testing, in some form | 24.7% |
| Stopping the AI doing something it shouldn't (guardrails) | 22.6% |
| Measuring whether the AI is actually any good (evaluation) | 18.8% |
| Accuracy, and catching made-up answers (hallucination) | 14.4% |
Those are testing instincts pointed at AI. And here's the tell: only 35 of the 2,110 JDs use the words "SDET" or "test automation." Nobody is advertising a "forward deployed tester" — the title doesn't exist — but the skill is wanted everywhere. That's the opportunity: the demand is real and unclaimed.
Dexity Intel · free newsletter
Liking this? Get the next one in your inbox.
JD-backed career reads, AI market signals, and field-tested tool guides — a few times a month. No fluff, no spam.
The catch: what testing doesn't teach you
A skill match is not the whole job, and it's important to be honest about the gap. A testing background covers part of what FDE JDs want — the quality, observability, and evaluation instincts. On top of that, an FDE still has to bring:
- A software-engineering base. Forward deployed engineering is built on engineering fundamentals; the AI layer sits on top of that, not instead of it.
- AI engineering. RAG, agents, evals-as-code, deployment — building AI systems, not just judging them.
- The customer-facing side. Deploying inside a client's environment and owning the outcome — something no testing job has ever asked of you.
That gap is exactly why the transition is a program, not a weekend of tutorials — and why the first question to ask yourself is about your coding experience, not your testing résumé.
How much more does forward-deployed engineering pay?
Enough to change the calculation. Forward-deployed roles run a median near $200,750 at mid level and $205,000 at senior, against $110K–$149K for a QA engineer and $73K–$96K for a quality analyst. Even the SDET, the best-paid testing role at $100K–$140K, sits well below the FDE median. The pay gap is the widest reason the move is worth the effort — but only if the engineering base is there to make it real.
Which testers should actually consider the move?
Be honest with yourself about two things — do you write code, and how senior are you:
- SDETs are the cleanest match in testing. You already build software; the jump is adding the AI engineering and the customer-facing layer.
- QA engineers who code and want a domain switch are strong candidates too — you're on the automation half of the role already.
- Quality analysts who write code can make it; those who don't can't jump straight to FDE — both FDE and machine learning are built on an engineering base. If you don't code, the honest path is AI skills you can use without programming, not a coding-heavy switch.
- The very senior (20+ years) face longer odds landing in a brand-new domain cold; the higher-probability path often keeps you in the domain you already own, with AI skills layered on.
FDE isn't the only route out of testing — but for a hands-on tester who wants a genuine domain switch and the pay that comes with it, it's one of the best-fitting the AI era has produced. The way to test the fit is to benchmark against the real skills, not the job title: walk through the FDE skills roadmap and ask which of those things you've built, not read about.
Frequently asked questions
What is the difference between a QA engineer and an SDET?
An SDET is a developer who builds the test automation (76% of SDET JDs require a programming language). A QA engineer owns quality for a product area — running and extending that automation, writing some tests, and still checking by hand — and only about half of QA-engineer JDs require coding.
What is a quality analyst versus a QA engineer?
A quality analyst checks software by hand against a requirements list and files defects, rarely writing code (a language appears in just 23% of their JDs). A QA engineer is more technical: they run and extend automated tests and own quality for part of the product.
Which testing role pays the most in 2026?
In Dexity's analysis of 1,243 US testing JDs, QA engineers top the range at $110K–$149K, just above SDETs at $100K–$140K, with quality analysts at $73K–$96K.
Can a QA engineer or SDET become a forward deployed engineer?
Yes — especially SDETs and coding QA engineers. FDE JDs are dominated by testing-adjacent skills (testing 24.7%, observability 30.8%, evals 18.8%, guardrails 22.6%), and forward-deployed roles pay a median near $200K. The gap to close is AI engineering and the customer-facing side, on top of a software-engineering base.
Is testing a good career in 2026?
It's under pressure from AI on the manual, checklist end — which is why nearly 6 in 10 testers Dexity spoke with want to move out, and 4 in 10 want to upskill in AI to stay relevant. The automation and quality-judgment skills, though, transfer unusually well into AI roles.
Related reading
- Forward Deployed Engineer — the complete 2026 guide — the role testing skills transfer into, in full.
- Forward Deployed Engineer Hiring Report — who's hiring, what they pay, and what the postings require.
- How to build a RAG system — one of the AI-engineering skills a tester would add on top.
Turn a testing background into a forward-deployed career
If you're a hands-on tester who wants out — and the pay gap says it's worth it — the move is to add the AI engineering and customer-facing skills on top of what testing already taught you. Dexity's Forward Deployed Engineering sprint builds exactly that: you ship a real AI workflow the way an FDE does, so you walk out with the artifact these roles screen for — proof you can make AI work, and judge whether it's good enough to ship.
Sources: Dexity's analysis of 1,243 US testing job descriptions and 2,110 forward-deployed-engineer JDs (2026), and Dexity's conversations with 80 testers. Figures are directional, US-only, from Dexity's own job-postings analysis and counseling data. · Dexity.com
Go from reading to doing · Dexity Course
Forward Deployed Engineering
The FDE Course starts straight at the distinctly-FDE skills — scope, build, deploy, and own real AI systems inside customer environments. Five weeks of craft plus a portfolio-grade capstone: no repetition of agent, RAG, or multi-agent foundations, and no interview prep.
