Building with the Claude API and agents
Code-heavy course in which you build AI agents yourself on the Anthropic API with tool use, MCP and agent patterns such as ReAct and plan-and-execute.
NRTO quality mark
Free cancellation up to four weeks before the start, so feel free to book now.
In brief. Building with the Claude API and agents is a 2-day classroom course. It is held in The Hague, in a group of no more than 8 participants. The price is EUR 1500 excl. VAT per participant, including course materials, lunch on site and a certificate of attendance. We also run this course for a team or department in-company.
- Build at least three working agents on the Anthropic API yourself
- Write your own MCP server that exposes your business tools to Claude
- Work in your own IDE with Python or TypeScript, against real endpoints
- Small groups of no more than eight participants and an experienced trainer
- Go home with code you can deploy directly in your own stack
You build agents yourself on the Anthropic API and learn how to connect Claude to real tools, databases and documents. From the first API call to a production agent with evaluation, logging and protection against prompt injection.
Description
The course starts with the basics of the Anthropic API. You learn the Messages API inside out: authentication, choosing between the Opus, Sonnet and Haiku models, streaming, vision, multi-turn conversations and system prompts. You understand how pricing works and how to make serious cost savings with prompt caching.
You then move on to tool use and document grounding. You define functions that Claude can genuinely call, connect it to existing APIs and databases, and build a Q&A bot over your own documents with citations. All in your own IDE, in Python or TypeScript.
In the second part you delve into agent patterns. You implement ReAct loops and plan-and-execute, distribute tasks across sub-agents and write your own MCP server (Model Context Protocol) that exposes your business tools to Claude. You learn how to evaluate agents with eval sets and scoring, and how to set up regression tests.
Finally, you cover everything you need for production: rate limits, fallbacks, error handling, logging, cost control and protection against prompt injection. You finish with a multi-step agent that carries out real tasks in your own context.
Topics
- The Anthropic API: authentication, model selection (Opus, Sonnet, Haiku) and pricing
- The Messages API: system prompts, roles and multi-turn conversations
- Streaming responses and vision (having Claude interpret images)
- Tool use: defining functions and having Claude call real APIs
- Document grounding with citations across your own documents
- Prompt caching to reduce costs with repeated context
- Agent patterns: ReAct loop, plan-and-execute and multi-step reasoning
- Model Context Protocol (MCP): when to use it and writing your own MCP server
- Sub-agents and delegation between specialised Claude instances
- Evaluation: eval sets, scoring and regression tests for your agent
- Production: rate limits, fallbacks, error handling, logging and cost control
- Security: mitigating prompt injection and validating output
Result
After the course, you will work independently with the Anthropic API and build AI applications that you can put straight into production. You know the Messages API inside out and know which model suits which situation. You have written agents using proven patterns and know how to measure their quality.
Specifically, by the end you will be able to:
- Building working integrations on the Anthropic API in Python or TypeScript
- Using tool use to connect Claude to APIs, databases and internal systems
- Applying document grounding and prompt caching for accurate and cost-efficient solutions
- Writing your own MCP server that exposes your business tools
- Designing agents using ReAct, plan-and-execute and sub-agent patterns
- Measuring the quality of agents with eval sets, scoring and regression tests
- Tackling production concerns: rate limits, fallbacks, logging and cost control
- Recognising and mitigating prompt injection
Target audience
This course is intended for backend developers, AI engineers, ML engineers and tech leads who want to integrate AI functionality into their own applications. You work with code on a daily basis and are comfortable with REST APIs, JSON and async patterns.
You have at least two years' experience in backend development in Python, TypeScript or a comparable language. Prior knowledge of AI or LLMs is not required. You learn the relevant concepts around Claude, tool use and agents during the course.
The course is technical and code-heavy. You write working agents yourself and go home with code that you can deploy directly in your own stack. Would you rather approach AI at a strategic level, or mainly learn to code with AI assistance? In that case, AI strategy for tech teams or Claude Code for developers suits you better.
Teaching method
This course is fully hands-on and code-heavy. You work in your own IDE with Python or TypeScript and build at least three working agents against the Anthropic API during the course.
Every module follows the same rhythm: a short explanation of the concept, a live demo by the trainer, and then you get to work yourself in the practice repository. You are given code templates that you adapt straight away to your own use case, so that by the end of the course you have code you can deploy in your organisation.
Two labs form the core: first you build a Q&A bot with document grounding, then a multi-step agent with tool use and evaluation. Because the group is small, with no more than 8 participants, there is plenty of time for code review and for discussing your specific stack with the trainer.
You take the course in a classroom on location or online. You receive API credits for the exercises.
Prior knowledge
You have at least two years' experience with backend development in Python, TypeScript or a comparable language. You are comfortable with REST APIs, JSON and async patterns. Prior knowledge of AI or LLMs is not required. You learn the concepts around Claude, tool use and agents during the course.
Comments
Bring your own laptop with a working Python or TypeScript environment. Arrange an Anthropic API key in advance, so that you can work against real endpoints from the very first moment.
Practical information
- Number of days
- 2 days
- Course times
- from 10 am to 5 pm
- Teaching hours
- 12 hours
- Education level
- Suitable for all levels
- Preparation time
- None
- Included
- Course materials and lunch on site
- Certificate
- At the end of the course you will receive a certificate of attendance
- Cancel
- Free up to four weeks before the start (cancellation terms)
Language
The course is given in Dutch as standard. The trainer speaks English. English-language course materials can be used. If at least 3 participants register, the course can also be given entirely in English.
In-company
Would you like to organise this training for a whole group or several groups? This can be cost-effective. For example, we can tailor the content entirely to your organisation, and training a whole group is also cheaper.
Are you looking at large numbers of participants or a complete academy? That involves a great deal. Consider planning and communication, for example, or identifying and agreeing the objectives and content of the training. Or differences in level, feedback mechanisms and safeguarding. Fortunately, Learnit has extensive experience in this field (more than 25 years) and we are happy to help you through this process.
Request a quote More about Building Claude API and agents in-company
All start dates
2 training days. In order of date.
- Thu 5 & Fri 6 November 2026 The Hague · 10:00–17:00 Enrol
- Thu 12 & Fri 13 November 2026 The Hague · 10:00–17:00Free cancellation until 15 October Enrol
No suitable start date? Send us a date request, and we'll look at the options together.
This course for your team?
Plan your own group on a date that suits you, at your office, at our premises or online. We tailor the content to your day-to-day work, and training a whole group is cheaper than individual bookings.
- Your date and location
- Tailored content
- Multiple groups or an academy are also possible
- More than 25 years' experience
What others say about this course
“Learnit is a professional organisation. The information is conveyed with enthusiasm.”
"Enjoyable and instructive course. A good balance between knowledge transfer and practical exercises. Very practically applicable in my own work."
"Very nice school to learn a variety of training with dedicated experts."
"Excellent"
"Good story, knowledgeable and inspiring, with a dose of humour too"
Training formats
Classroom
2 days€ 1500,-- Pleasant venue with a good trainer
- Small groups
- Inspiring and motivating
In-company
Bespoke trainingOn request- 100% for your group
- Bespoke training
- At your place or ours
Workshop
Up to 4 hours for groupsOn request- Suitable for events
- Interactive and inspiring
- Large groups too
Private training
At our place or yoursOn request- 100% just for you
- Achieve a lot in little time
- Whenever you want
Frequently asked questions
Can I cancel free of charge?
Yes. Up to four weeks before the first day of the course you can cancel free of charge, without giving a reason. You then receive a credit note or, if you have already paid, a refund. If you cancel later, you pay part of the course fee: 25% between four and two weeks before the start, 75% in the two weeks before it and 100% in the final week. Moving to another start date is free of charge up to two weeks before the start, after that 50%. Sending a colleague in your place is always possible. Read the cancellation terms.
What does the Claude API and building agents course cost?
You can find the current price at the top right of this course page, along with the schedule, location and delivery format. For in-company or bespoke courses we are happy to provide a quotation on request.
What prior knowledge do I need for the Claude API and building agents course?
You have at least two years' experience with backend development in Python, TypeScript or a comparable language. You are comfortable with REST APIs, JSON and async patterns. Knowledge of AI or LLMs is not required. You learn the concepts around Claude, tool use and agents during the course itself, with plenty of practical examples.
Who is the Claude API and building agents course intended for?
The course is intended for backend developers, AI and ML engineers, tech leads and solution architects who want to build AI functionality into their own applications. If you would rather approach AI strategically, our AI strategy course is a better fit. If you are mainly looking to code more productively with AI assistance, take a look at Claude Code for developers.
What exactly will you learn during the Claude API and building agents course?
You learn to work with the Anthropic API: the Messages API, choosing between the Opus, Sonnet and Haiku models, streaming, vision, tool use, document grounding with citations and prompt caching. You also build agents using patterns such as ReAct and plan-and-execute, write your own MCP server, work with sub-agents and learn to evaluate, monitor and protect against prompt injection.
In which delivery formats can I take the Claude API and building agents course?
You can take the course in a classroom at one of our locations, in a virtual classroom via video conferencing, or in-company at your own office. For in-company courses we tailor the use cases and examples to your own stack and business context. The current schedule and delivery formats can be found on the course page.
Do I receive a certificate after the Claude API and Building Agents course?
Yes, on completion you receive a certificate of participation from Learnit. This shows that you have mastered the Anthropic API, tool use, MCP and agent patterns, and that you can build production-grade Claude agents independently.
Which teaching methods and materials are used in the Claude API and Building Agents course?
The course is hands-on and code-heavy. You work in your own IDE with Python or TypeScript and build at least three working agents. Each module consists of a short explanation, a live demo and coding yourself in the practice repository. You receive code templates, a course pack with reference material and access to a fully worked example repository that you keep afterwards.
Why choose Claude instead of OpenAI or open-source LLMs?
Claude excels at tool use, long context, document grounding with citations and structured reasoning for agents. With MCP, Anthropic has established an open standard that is rapidly being adopted. In the course, however, you learn general agent patterns that you can largely apply to other LLMs as well. The focus is on Claude because that stack is currently the most mature for production agents.
Can I apply what I have learned directly in my own projects?
Yes, that is the idea. During the course you work in your own IDE, in the programming language you use every day, against real Anthropic endpoints. You adapt the code templates you receive directly to your own use case, so that by the end you have working agents you can roll out in your organisation. Feel free to bring along a concrete problem from your backlog.
What is the difference with the Claude Code for Developers course?
Claude Code for Developers focuses on coding more productively with AI assistance in your own IDE, using tools such as Claude Code as a pair programmer. This course focuses on something else: you build AI functionality into your own applications, via the Anthropic API, tool use, MCP and agent patterns. The first makes you as a developer faster. This one makes your software smarter.
I am a tech lead and not a full-time developer. Is the Claude API and building agents course still suitable for me?
Yes, provided you still have enough affinity with code to take part in the labs yourself. Many tech leads take the course precisely in order to make well-founded architectural choices around LLM integrations, MCP and agents. You will go home with a solid picture of what does and does not work in production, and you will be better able to steer your team on agent projects.
How do you prevent prompt injection and other security risks in the agents you build?
Prompt injection is one of the biggest risks with agents that use tools, certainly if you load user input or external documents into the context. In the course we cover concrete mitigations: input sanitisation, separation of instructions and data, output validation, least privilege in tool definitions, allow lists for sensitive actions and logging tool calls for auditing. You practise with these directly in your own agents.
Next start date: Thu 5 & Fri 6 November 2026 in The Hague
Free cancellation up to four weeks before the start, so you can secure a place now.