Last updated October 2026.
Three years ago, searching a job board for "AI" mostly turned up machine learning engineers and data scientists. In 2026 you find a dozen new titles, several that mean the same thing, and a few that mean something different at every company that uses them.
This guide covers ten roles that have become real categories: what each does, who hires, which backgrounds transfer, and which alternative titles to search so you don't miss half the postings.
Why the Titles Are Such a Mess
The growth is real. LinkedIn's 2026 Jobs on the Rise list put AI engineer as the fastest-growing job title in the US, with postings up 143% year over year, and the share of US postings asking for "agentic AI" skills went from 0.06% to 0.23% in a single year, roughly a 280% jump (JobAgent's summary of the data).
But the vocabulary hasn't settled. Companies invent titles to look current, and the same work gets labeled three ways depending on who owns the budget. Ivan Turkovic's breakdown of AI job title naming chaos explains how this happened. These jobs aren't only in engineering: roughly half of AI postings sit outside IT departments, according to this analysis of the 2026 AI jobs market.
The practical takeaway: read the responsibilities, not the title.
The 10 Roles
1. AI Engineer
What they do: Build product features on top of large language models: retrieval, prompting, tool calls, evaluation, and the plumbing between a model and real data.
Who hires: Almost everyone shipping software, from startups to banks. It's the umbrella title, and has largely absorbed "GenAI engineer."
Background that transfers: Backend or full-stack engineering is the most common path. No research background needed.
What to put on your resume: Production features that use a model, with outcomes. Name what you built (retrieval pipeline, eval suite, cost work), not just the model you called.
Search terms: AI Engineer, GenAI Engineer, LLM Engineer, Applied AI Engineer, AI Software Engineer.
2. Forward Deployed Engineer
What they do: Engineers who sit with customers and build whatever makes an AI product actually work inside that customer's systems.
Who hires: AI labs, AI infrastructure companies, and enterprise AI startups selling to large customers. Indeed postings for the role rose about 729% year over year from April 2025 to April 2026 (Paraform).
Background that transfers: Software engineers who like customer contact, and solutions engineers or consultants who can really code.
What to put on your resume: Customer-facing builds: what you shipped, for whom, and what it changed. More in the forward deployed engineer career guide.
Search terms: Forward Deployed Engineer, FDE, Deployment Engineer, Customer Engineer (AI), Field Engineer.
3. AI Agent Engineer
What they do: Build software that loops: the model plans a step, calls a tool, reads the result, and decides what to do next until a task is done. The hard parts are reliability, permissions, and knowing when to stop and ask a human.
Who hires: Companies automating multi-step work like support, operations, and coding.
Background that transfers: Backend engineers who know workflows, queues, and observability.
What to put on your resume: Agents or automations you shipped, the tools they could call, how you measured success, and how you handled failures.
Search terms: AI Agent Engineer, Agentic AI Engineer, Agent Developer, AI Automation Engineer, AI Engineer (Agents).
4. AI Evals Engineer / AI Quality Engineer
What they do: Measure whether an AI system is any good: test sets, scoring rubrics, automated graders, and regression checks that show whether a change made things better or worse.
Who hires: AI labs and any company with a serious AI product. Evals is pulling apart as its own category, but most postings are still labeled AI Ops or MLOps (JobsByCulture's evals career guide).
Background that transfers: QA and test automation, data science, and analytics.
What to put on your resume: Eval sets you built, metrics you defined, and a case where your evaluation caught a regression or changed a decision.
Search terms: AI Evals Engineer, AI Quality Engineer, LLM Evaluation Engineer, AI Ops Engineer, MLOps Engineer, Model Evaluation.
5. Context Engineer
What they do: Decide what information a model sees for each request: which documents, history, and instructions, in what order and size.
Who hires: Mostly AI-native startups and agent teams. It's the newest title here, appearing in late 2025.
Background that transfers: Search and retrieval engineering, information architecture, technical writing paired with programming.
What to put on your resume: Retrieval or context work you did and what it improved: accuracy, cost, or response quality.
A warning: Some "context engineer" postings are relabeled prompt-writing or consulting work. Ask in the screen what the role actually ships.
Search terms: Context Engineer, Retrieval Engineer, RAG Engineer, AI Engineer (Retrieval), Prompt and Context Engineer.
6. AI Solutions Architect
What they do: Design how AI fits into existing systems: which models, where data lives, how security and cost are handled. More design and pre-sales than building.
Who hires: Cloud providers, AI vendors, consultancies, and large enterprises with internal architecture teams.
Background that transfers: Solutions architects, cloud architects, and senior engineers who already do customer-facing design work.
What to put on your resume: Architectures you designed and their results. Real deployments beat certifications.
Search terms: AI Solutions Architect, GenAI Solutions Architect, AI Architect, Solutions Engineer (AI), AI Technical Consultant.
7. AI Product Manager
What they do: Own an AI feature: what it does, how quality is judged, what it costs, and what happens when the model is wrong.
Who hires: Software companies of every size, plus non-tech companies building internal tools.
Background that transfers: Product managers with data or technical depth, and engineers or analysts moving into product.
What to put on your resume: AI features you shipped, how you defined success, and the quality-cost-speed tradeoffs you made.
Search terms: AI Product Manager, Product Manager (AI/ML), GenAI Product Manager, Technical Product Manager (AI), Product Lead, AI.
8. AI Governance / Responsible AI Lead
What they do: Set the rules for how a company builds and uses AI: risk reviews, policy, compliance, and bias and safety checks.
Who hires: Regulated industries, large enterprises, and AI labs, often in legal or risk rather than engineering.
Background that transfers: Privacy, compliance, risk management, legal, and audit. A technical background helps but isn't always required.
What to put on your resume: Policies, review processes, and audits you created or led.
Search terms: AI Governance Lead, Responsible AI Lead, AI Risk Manager, AI Ethics Lead, AI Compliance Manager, AI Policy.
9. AI Trainer / Data Specialist
What they do: Write, grade, and correct model output so models learn what good answers look like, much of it feedback for reinforcement learning (RLHF). Increasingly the work goes to domain experts: lawyers, doctors, and programmers paid to judge answers in their own field.
Who hires: AI labs, directly or through data and annotation vendors. A lot of the work is contract or part-time.
Background that transfers: Deep expertise in almost any field, plus careful writing. Here a professional credential matters more than coding.
What to put on your resume: Your domain credential up top, plus any annotation or review work.
Search terms: AI Trainer, AI Data Specialist, RLHF Specialist, AI Tutor, Expert Annotator, AI Writing Evaluator, plus your field (for example "AI Trainer legal").
10. AI Enablement / AI Adoption Lead
What they do: An internal role that rolls out AI tools across a company: picking tools, training teams, and measuring whether work actually got faster.
Who hires: Mid-size and large companies, often in operations, IT, or HR rather than engineering.
Background that transfers: Change management, IT program management, and learning and development.
What to put on your resume: Rollouts you led, adoption numbers, and measured time saved.
Search terms: AI Enablement Lead, AI Adoption Manager, AI Transformation Lead, AI Program Manager, Head of AI Enablement.
Quick Summary Table
| Role | Best-fit background | Coding required? |
|---|---|---|
| AI Engineer | Backend / full-stack engineering | Yes |
| Forward Deployed Engineer | Customer-facing software engineering | Yes |
| AI Agent Engineer | Backend, workflows, distributed systems | Yes |
| AI Evals / Quality Engineer | QA, test automation, data science | Yes, mostly |
| Context Engineer | Search, retrieval, information architecture | Usually |
| AI Solutions Architect | Solutions or cloud architecture | Some |
| AI Product Manager | Product management with technical depth | No, but helps |
| AI Governance / Responsible AI | Compliance, risk, legal, privacy | No |
| AI Trainer / Data Specialist | Any deep domain expertise | Only for coding tracks |
| AI Enablement / Adoption | Change management, IT programs, ops | No |
Titles That Are Fading
Prompt Engineer as a standalone title is the clearest example. In 2023 it was the headline AI job. By 2026 it shows up far less on its own, because the skill got folded into every role on this list (ilinmaks.com).
If prompt engineering is your main experience, reframe it under the role you're targeting: "built and evaluated prompts for a support agent" reads as AI engineering or evals work.
How Naming Chaos Affects Your Job Search
You miss postings. Search only "AI Engineer" and you won't see the same job posted as "LLM Engineer." Save a search for each title in the search-terms lines above.
Your resume gets filtered on the wrong word. ATS software and recruiters both match on titles. If the posting says "AI Agent Engineer" and your headline says "Software Engineer," you look like a weaker match even when your experience fits. So:
- Mirror the posting's exact title in your resume headline, as long as it honestly describes what you do.
- Use the posting's vocabulary in your bullets. If they say "evaluation," don't only say "testing."
- Keep past job titles accurate. Change the headline and summary, never your title history. Our guide on how to beat ATS in 2026 covers what matching does and doesn't reward.
More titles means more postings and more resume versions. That's worth it only if the repetitive parts don't eat your week.
Where ProfVault Fits
ProfVault keeps one structured profile (jobs, dates, education, skills, standard answers) and uses it for every application. The Chrome extension fills Workday, Greenhouse, Lever, Ashby and other ATS forms from that profile, including dropdowns, and can attach your resume file, free and with no limit on applications. For open questions like "Why this role?", it can draft an AI-written answer from your profile on 10 applications a month for free. You review every field before you submit.
New accounts start with a 7-day Pro trial: unlimited AI answers in the extension and 5 AI credits for tailored resumes and cover letters, which is useful for matching your headline and bullets to a posting's exact title. If you're applying across several AI titles this week, that's the window to set up your profile. It pairs well with a one-week job sprint.
FAQ
What is the difference between an AI engineer and an ML engineer?
ML engineers usually train and deploy models. AI engineers usually build products on top of existing models, using retrieval, prompting, tools, and evaluation. The line is blurry, and many postings use both titles.
Is "prompt engineer" still a real job?
It still appears, but rarely as a standalone title. The skill is now expected inside other AI roles, so present it as part of engineering, evals, or product work.
How do I know if a "context engineer" posting is a real engineering role?
Check what the role ships. Real context engineering involves building retrieval and context pipelines and measuring their effect. If the responsibilities are mostly writing prompts for clients, it's consulting under a new name.
Related reading
- Forward Deployed Engineer: the 2026 career guide
- What AI companies look for when hiring in 2026
- Best ATS-Friendly Resume Format in 2026
- The one-week job sprint: 50 applications, 10 truly tailored
Ready to stop retyping your resume into every job application? Install the ProfVault Chrome extension (free) and auto-fill any ATS in one click.