Role hub · live · updated 20 Nov 2026

AI engineer: the role, pay and who is hiring

An AI engineer builds products and features on top of AI models: picking the model, wiring it to the company’s data, writing the prompts, tools and evaluations, and running it in production. It is the fastest-growing engineering title of the last two years, hired by AI startups, banks, consultancies and big tech alike.

Open roles
1,469
Companies hiring
717
Median posted pay
$150k to $213k
Remote
21%
Demand
22k searches a month, up 1.6x

This week · 20 Nov 202675 new AI engineer roles, most from Capco, RVO Health and IFS.

01 · The role

What does an AI engineer do

Less model training, more shipping. Read across live postings, the work splits into four parts:

  1. BUILD

    Ship the AI feature

    Turn a product idea into a working feature: model choice, prompts, tool calls, retrieval over the company’s data, and the API around it.

  2. MEASURE

    Write the evals

    Build test sets and automatic checks, so a prompt or model change is judged on numbers rather than a demo.

  3. RUN

    Keep it fast and cheap

    Latency, cost per call, caching, fallbacks between models, and monitoring for bad answers in production.

  4. IMPROVE

    Close the loop

    Use real usage and failures to improve prompts, data and, when it pays off, fine-tuned models.

What AI engineer postings ask for

  • LLMs 32%
  • RAG 25%
  • Agents 15%

Share of 2,084 live AI engineer postings naming each skill, read 19 Nov 2026. ML engineer postings, for contrast: PyTorch 37%, TensorFlow 23%, Spark 12%.

The clean test: an ML engineer improves the model, an AI engineer improves the product built on it. Read the long version →

02 · Pay

AI engineer salary

Median posted pay on landed, next to the bands the biggest employers publish.

Median posted range, USD base

$150k to $213k

From 340 live postings that publish a range, before equity and bonus.

AI engineer salary by city and company →

03 · The way in

How to become an AI engineer

Most AI engineers were software engineers first. The way in, as a checklist:

  1. [x]
    Be a solid software engineerAPIs, databases, deploys. AI features are still software.
  2. [x]
    Ship one real LLM featureRetrieval over real documents or an agent with tools, used by someone other than you.
  3. [ ]
    Learn evalsBuild a test set and an automatic grader for your feature. It is what interviewers probe most.
  4. [ ]
    Know the cost sideToken costs, caching, smaller models, latency budgets. Companies hire for this.
  5. [ ]
    Apply the day roles openEarly applicants get read first. Full AI engineer roadmap →

04 · The roadmap

How to become an AI Engineer: the roadmap

About Depends on starting point, 5 stages, 6 core skills, aiming at See salary data. There is no universal number of months required to become an AI Engineer. Start from your current engineering foundation, then build evidence across software delivery, model-backed applications, evaluation, data and security boundaries, and production operations. The finish line is not a course. It is one system you can explain, test, and improve.

  1. Stage 0

    Map the target role

    Compare three current job descriptions and identify repeated outcomes, constraints, and evidence

  2. Stage 1

    Software foundations

    Python or TypeScript, APIs, data, tests, debugging, Git, and deployment

  3. Stage 2

    Model-backed applications

    Structured outputs, retrieval, tools, model choice, and failure handling

  4. Stage 3

    Evaluation and safety

    Representative tasks, graders, error analysis, permissions, and prompt-injection defenses

  5. Stage 4

    Production evidence

    Ship one observable system with users, traces, cost, latency, and a written postmortem

The full roadmap: projects, skills and what to learn at each stage

What does an AI Engineer own?

The title varies. In current practice, an AI Engineer usually turns models into a useful, reliable product or workflow. That can include application code, retrieval, tools, evaluations, data access, model selection, observability, cost, and incident handling.

Do not choose a curriculum from the title alone. Compare three current postings in the same market and company type. Write down the repeated outcomes and the evidence each employer asks for.

Stage 0: map the target role

Build a requirement-to-evidence table before studying:

Repeated requirementEvidence you already haveGap to close
Build production servicesShipped API or applicationAdd reliability and operational ownership
Work with model-backed featuresPrototype or experimentAdd evaluations and real users
Retrieval or toolsSearch, data, or integration workBuild and measure one system
Security and data boundariesAuthorization experienceApply it explicitly to model and tool access

This prevents an endless tour of frameworks that the target employers never mention.

Stage 1: strengthen software foundations

You need one production language, APIs, data modeling, tests, debugging, Git, and deployment. Python is common in model and data work; TypeScript is common in product surfaces. Use the language that lets you ship the target system well.

Learn enough machine learning to reason about model behavior and evaluation. Research-heavy roles require much deeper mathematics and training knowledge than applied-product roles. The job description decides the depth.

Stage 2: build model-backed applications

Start with the smallest system that exposes real engineering decisions:

  • structured inputs and outputs;
  • a model call with retries and error handling;
  • retrieval only when the task needs external information;
  • tools with narrow permissions;
  • logs that show latency, cost, and failures;
  • a small evaluation set tied to the user task.

Framework choice is secondary. You should be able to explain the architecture without naming a framework.

Stage 3: learn evaluation and safety

Anthropic's agent-evaluation guide recommends matching the evaluation method to the task rather than relying on one score. Build representative cases, inspect individual failures, and combine deterministic checks, human judgment, and model graders where each is valid.

Treat authorization as code, not as a prompt. Test tenant isolation, tool permissions, prompt injection, malformed inputs, and partial failure.

Stage 4: ship one strong portfolio system

A credible project has:

  1. a real user or workflow;
  2. a written objective and failure taxonomy;
  3. a representative evaluation set;
  4. observable traces, latency, and cost;
  5. permission boundaries for tools and data;
  6. a postmortem describing what failed and what changed.

One measured system is better evidence than five tutorial clones.

How do you prepare for interviews?

Official guidance from OpenAI and Anthropic still emphasizes coding, testing, debugging, tradeoffs, communication, and role-specific skill. Use the AI Engineer interview question bank to practice those decisions without pretending that every company uses one universal loop.

How long will it take?

It depends on the starting point and target. A backend engineer who has shipped services needs different work from a researcher who has not owned production, or a beginner who is learning programming. Set evidence gates instead of month promises:

  • Can I ship and debug an API?
  • Can I explain why a model-backed feature fails?
  • Can I build a valid evaluation set?
  • Can I enforce data and tool permissions?
  • Can I show a deployed system and a postmortem?

When those answers are concrete, compare them with current job descriptions and apply selectively.

05 · The difference

AI engineer vs ML engineer vs software engineer

The titles overlap. What differs is what the role is measured on.

Software engineerML engineerAI engineer
Works onProduct features and systemsModels: training, features, servingProducts built on existing models
Main toolsLanguages, databases, cloudPyTorch, data pipelines, GPUsModel APIs, retrieval, agents, evals
Measured byProduct metrics, reliabilityModel accuracy and costAnswer quality, latency, cost per call
Typical backgroundCS or bootcampML or statistics heavySoftware engineer who learned LLMs

06 · The interview

The AI engineer interview

Most loops keep the usual coding round and add one or two rounds about building with models.

  1. Coding

    A standard coding round, often in Python, sometimes on an API or data task rather than a puzzle.

    Most companies
  2. LLM system design

    Design a feature such as a support agent or document search: retrieval, prompts, tools, evals, failure handling.

    The core round
  3. Practical build

    A take-home or live build with a model API, judged on how you measured quality, not just that it runs.

    Startups often
  4. Past work

    Walk through something you shipped with AI: what broke, how you measured it, what it cost.

    Every loop
The full AI engineer interview guide →

07 · Hiring

Who is hiring AI engineers

1,469 open roles at 717 companies, refreshed daily. Each links to the company's own page.

09 · Hiring now

AI engineer jobs

1,469 open today. The newest openings first, from the companies' own career pages.

Get your AI engineer applications out the day roles open.

landed watches every AI engineer opening and applies within the hour, with answers written for each company. 10 free applications, no card.

Start applying free →

FAQ

AI engineer, answered

What is a ai engineer?

An AI engineer builds products and features on top of AI models: picking the model, wiring it to the company’s data, writing the prompts, tools and evaluations, and running it in production.

How much does a ai engineer earn?

Across 340 live AI engineer postings on landed that publish pay, the median range is $150k to $213k in USD base, before equity and bonus. Senior roles at AI labs post well above that.

Who is hiring AI engineers?

landed tracks 1,469 live AI engineer roles at 717 companies. Capital One, Micron, Kyndryl, Devoteam, Capco, Booz Allen Hamilton, Mastercard and HP Enterprise have the most open right now.

Are AI engineer jobs remote?

21% of live AI engineer postings on landed are fully remote. Most of the rest are hybrid in the biggest tech cities.

What is the difference between an AI engineer and an ML engineer?

An ML engineer trains, tunes and serves models. An AI engineer builds products on top of existing models: retrieval, agents, prompts and evaluations. Many teams have both; small startups often hire one person who does a bit of each.

How do I become an AI engineer?

Most come from software engineering. Ship one real LLM feature that others use, learn to measure its quality with evals, and understand latency and cost. A degree in ML helps but is rarely required.