The New AI Engineering Landscape
The AI wave didn't just create new tools — it created new jobs that didn't exist 3 years ago. Find where you fit.
Two Axes, One Map
Where you sit on these two axes defines which role fits you.
Closer to the model
You work with model internals — training, fine-tuning, evaluation, safety. High barrier to entry, fewer roles, highest research overlap.
Closer to production
You build systems that ship. Models are a tool, not the subject. Highest job volume, fastest growth, most accessible from traditional engineering.
Most new job demand sits in the production-facing quadrant. The applied, product-facing roles are where the 2025–2026 growth is.
13 Roles, Mapped
From well-established to just emerging. Each card links to the full profile in the guide.
Prompt Engineer
EstablishedCraft and optimize instructions sent to AI models for reliable, high-quality outputs.
Context Engineer
GrowingDesign systems that give AI models the right information, at the right time, in the right format.
AI Engineer
GrowingBuild end-to-end AI-powered applications — from LLM integration to production monitoring.
LLM Engineer
GrowingDeep specialization in LLM integration, fine-tuning, and evaluation infrastructure.
AI Agent Engineer
GrowingDesign and build autonomous agent systems that plan, reason, and execute multi-step tasks.
Founding AI Engineer
GrowingBuild the AI core of an early-stage company end-to-end, from architecture to customer interaction.
AI Architect
EstablishedDesign enterprise AI systems — technology choices, reference architectures, governance standards.
Platform Engineer
EstablishedBuild the internal platform that makes AI development reliable, secure, and consistent across teams.
Harness Engineer
EmergingBuild the infrastructure that keeps AI agents under control — architectural constraints, context systems, and entropy management.
AI Product Manager
GrowingOwn AI-powered products — requirements, quality tradeoffs, UX of non-deterministic systems.
AI Safety & Eval Engineer
GrowingEnsure AI systems behave safely and reliably — build eval pipelines, red-team models, implement guardrails.
ML Engineer
EstablishedDevelop, train, deploy, and maintain machine learning models — the most traditional AI engineering role.
Career Decision Matrix
Where to go from where you are now.
| Your current profile | Best next role | Timeline |
|---|---|---|
| Software engineer (3+ years) | AI Engineer | 3–6 months upskill |
| Software engineer at early startup | Founding AI Engineer | Now, if opportunity exists |
| Backend engineer, infra-minded | Platform Engineer (AI) | 6–12 months |
| Senior engineer who thinks in systems | AI Architect or Harness Engineer | 1–2 years accumulation |
| Engineer who likes research & rigor | LLM Engineer or AI Safety/Eval | + ML foundations needed |
| Non-technical, works with AI daily | Prompt → Context Engineer | 6–18 months |
| PM who wants to stay PM | AI Product Manager | 3–6 months upskill |
| Engineer obsessed with reliability & archi | Harness Engineer | Pioneers' territory |
The fastest path to AI employment in 2025–2026
- Build something with AI APIs — a real project, not a tutorial
- Write about it — blog post, GitHub README, LinkedIn thread
- Add evals — measure your system's quality, show the numbers
- Apply for AI Engineer roles — the bar is demonstrated building, not credentials
76% of candidates claiming AI expertise lack production-level deployment experience. The bar is lower than it looks if you've actually shipped something.
Salary Benchmarks
Source: FinalRoundAI, Alcor, RiseWorks AI Talent Report (2025)
| Role | Status | Entry | Mid | Senior | Notes |
|---|---|---|---|---|---|
| Prompt Engineer | Established | $80K–$110K | $110K–$150K | $150K–$180K | Shrinking standalone market |
| Context Engineer | Growing | $100K–$140K | $140K–$180K | $180K–$230K | |
| AI Engineer | Growing | $120K–$160K | $160K–$220K | $220K–$300K | |
| LLM Engineer | Growing | $130K–$170K | $170K–$250K | $250K–$350K | Lab-level roles higher |
| AI Agent Engineer | Growing | $130K–$170K | $170K–$240K | $240K–$320K | |
| Founding AI Engineer | Growing | $100K–$150K + equity | — | — | Equity makes total comp wide-ranging |
| AI Architect | Established | — | $180K–$260K | $260K–$380K | |
| Platform Engineer | Established | $110K–$150K | $150K–$210K | $210K–$280K | |
| Harness Engineer | Emerging | — | — | — | Role emerging — no data yet |
| AI Product Manager | Growing | $130K–$170K | $170K–$230K | $230K–$350K | FAANG premium significant |
| AI Safety & Eval Engineer | Growing | $140K–$180K | $180K–$250K | $250K–$400K | Lab compensation highest |
| ML Engineer | Established | $100K–$140K | $140K–$200K | $200K–$280K |
What's Not a Role (Yet)
Terms you'll hear that describe practices, not job titles.
"Vibe Coder"
A methodology (use AI to handle implementation while you focus on design), not a job. Andrej Karpathy coined the term then moved to "context engineering" as more precise. No serious company has "Vibe Coder" on a job description.
"AI-Native Engineer"
Describes a quality expected of all engineers increasingly — you use AI tools fluently in your daily workflow. It's the bar, not the title.
"Orchestration Engineer"
Sometimes used for agent systems. Overlaps significantly with AI Agent Engineer. Not yet a distinct category in job postings.
Full profiles in the guide
Each role has a complete profile: detailed responsibilities, required skills, entry paths, and where the role is heading. Free, open-source.