Anthropic's coding agent for the terminal, your IDE and a desktop app, included in paid Claude plans.
Paid
AI engineer
New roleA developer who builds products and internal tools on top of language models: document assistants, agents, systems that handle incoming requests. On LinkedIn's Jobs on the Rise 2026 list for the US, this job ranks first.
Where this job sits on the map
In the same group: 14 of 100 professions.
What changes
You usually use a ready-made model through an API (a way for software to connect to it), and the work is everything around it: data, tools, checks, interface.
An agent writes most of the code, while the engineer owns the architecture, the quality of the answers and the cost of each request.
There are new duties: tests for answer quality (evals) and defense against prompt injection, where instructions get slipped in through incoming text.
AI drafts, the person decides
How the work splits here: AI prepares a draft, the person checks it and makes the call.
What AI does
- Writes the code that connects to the model and the plumbing around it
- Builds sets of test examples and reruns them after every change
- Reads request logs and finds where the assistant goes wrong
- Prepares documentation and change notes
You check itRead it, check the facts, fix it. Without this check the draft goes nowhere.
What stays with the person
- Choosing how the system is built: where one request is enough, where you need search over documents, and where you need an agent
- Security: what data and actions the model can reach
- Deciding that the quality is good enough to launch
- Telling the client plainly what AI can and can't do
What to learn first
6 lessons from the course, in order. Start with the first one.
- How an LLM works inside, explained without mathStarting from zeroHow a language model works inside: tokens, context, temperature and hallucinations, explained without math.Start here
- How to write a good promptUserHow to give AI a task: the five parts of a good prompt, Plan Mode, and how to refine an answer instead of starting over.
- APIs and integrationsBuilderWhat an API is: requests, keys, response codes and first integrations with Stripe, Twilio and SendGrid.
- RAG: Retrieval Augmented GenerationBuilderRAG: how to teach Claude to answer from your own documents using embeddings and a vector database.
- Evals: skills that improve themselvesEngineerEvals for skills: tests, pass rate, checking that the skill triggers and a data-driven improvement loop.
- Prompt injection defense, the top security threat of 2026EngineerPrompt injection: six attack types, eight layers of defense, red team tests and an incident response plan.
Which tools to use
A VS Code-based code editor with AI agents, a free Hobby plan and Pro at $20 a month.
FreemiumOpenAI's coding agent: parallel background tasks, a desktop app and the cloud.
FreemiumGoogle's free studio for Gemini: Build mode creates an app from a description or sketch.
FreemiumTurns a web page into clean text or a table for AI models and automations.
Freemium
Ready-to-use materials
- CLAUDE.md for a business projectCLAUDE.md
An instruction file for Claude Code in your business documents folder, no code involved.
Free Task, context, limits and checks, so the agent does the right thing and breaks nothing.
Free- Data security when working with AIChecklist
What not to paste into a chat, which settings to check and how to limit AI agents.
Free
How to earn with AI
We don't promise income: results depend on your niche, your market and your work.
- Job
If you're a developer, take on model-related work on your own team: a knowledge-base assistant, document processing, answer-quality checks.
- Service
A turnkey assistant for a company's documents: connect the sources, test the quality, hand it to the team with instructions.
- Product
A small service built on a model for a niche you know: sorting incoming requests, template-based reports, catalog search.
Similar professions
Checked: October 2026