ON
← Back to feed
Last Call: Claude Code in the field Develop AI chat agents in 5 sessions
Germany💻 Technology11 days ago

Last Call: Claude Code in the field Develop AI chat agents in 5 sessions

The article promotes a training course titled 'Agentische Entwicklung – mit Claude Code, Mastra und CopilotKit zum eigenen KI-Chat-Agenten' offered by heise Academy. The course teaches participants how to develop their own AI chat agents using tools like Claude Code, Mastra, and CopilotKit. Over five sessions, attendees will build a complete chat agent through a practical example project, covering topics such as integrating coding agents into development environments, implementing secure tool calls, and deploying applications with automated quality assurance. The curriculum includes hands-on experience with Git workflows, security measures against prompt injection, and building user interfaces with generative UI components. The course concludes with the deployment of the chat agent using CI/CD pipelines. Participants can access additional resources through the heise Academy platform.

A new training program titled Last Call: Claude Code in der Praxis, KI-Chat-Agenten in 5 Sessions entwickeln has been announced by heise academy, offering participants the opportunity to develop AI-powered chat agents using Claude Code, a product developed by Anthropic. The course will run over five sessions, each lasting four hours, with classes scheduled from September 26 to October 29, starting at 9 AM and ending at 1 PM daily. Participants will work alongside expert Rainer Stropek to build a complete chat agent through a specific project example, learning how to integrate the tool into terminal environments, IDEs, desktop applications, and CI/CD pipelines. The program aims to guide participants through the entire development process, beginning with setting up a project structure and moving through implementing custom tool calls with typed functions and Zod schemas. Security measures such as unit tests, integration tests, and end-to-end testing will be emphasized, along with the use of Git Worktrees to explore multiple design variations without conflicts. A dedicated section will focus on securing the agent through permissions, approval strategies, and safeguards against prompt injection and tool poisoning, both for the Claude Code harness and for the participant’s own agent. Later sessions will involve building a fully functional agent interface using CopilotKit and the open AG-UI protocol, incorporating features such as streaming, shared state, and generative user interfaces. Automation tools like skills and hooks will streamline repetitive tasks such as formatting, testing, and secret scans. The final session will cover deploying the chat agent in production, including CI/CD processes via GitHub Actions, automated reviews using sub-agents, and implementing a clean human-in-the-loop model. Participants can begin their journey with the first session on September 26, which will introduce them to the landscape of coding agents and help them set up their first project. The second session on September 8 will focus on using coding agents locally, exploring Git workflows and worktrees. On September 15, the emphasis will shift toward securely connecting to an MCP server and safeguarding both the Claude Code harness and the participant's own agent. The fourth session on September 22 will delve into orchestrating Claude Code, applying generative UI techniques, and using hooks in practice. Finally, on September 29, the focus will be on the agent SDK, sub-agents, and making the agent production-ready through CI/CD processes. The program is designed to be adaptable, allowing participants who bring their own domain knowledge to apply the concepts directly without requiring changes to the curriculum. This flexibility enables individuals to create specialized agents such as internal research assistants, DevOps insights agents, or support copilots tailored to their specific knowledge bases. The sessions will include extensive hands-on experience, interactive elements, and opportunities for participants to review recorded materials and exercises after the live sessions. Questions will be addressed directly during the live chat, and participants will have the chance to engage in discussions with one another throughout the course. Further information and registration details are available on the classroom website. Those interested in attending can access the site to learn more about the program and secure their spots. The course is part of heise academy’s broader offerings, which provide practical and expert-led training in various technological fields.

Go to the primary sources (2)

The official sources this coverage is built on. Read them directly to bypass framing.

1 reports

heise online logoheise onlineIndependentCenterFactual 94Objective 9711 days ago
Last Call: Claude Code in the field Develop AI chat agents in 5 sessions

The article promotes a training course titled 'Agentische Entwicklung – mit Claude Code, Mastra und CopilotKit zum eigenen KI-Chat-Agenten' offered by heise Academy. The course teaches participants how to develop their own AI chat agents using tools like Claude Code, Mastra, and CopilotKit. Over five sessions, attendees will build a complete chat agent through a practical example project, covering topics such as integrating coding agents into development environments, implementing secure tool calls, and deploying applications with automated quality assurance. The curriculum includes hands-on experience with Git workflows, security measures against prompt injection, and building user interfaces with generative UI components. The course concludes with the deployment of the chat agent using CI/CD pipelines. Participants can access additional resources through the heise Academy platform.

Bias read (Center): The article focuses on a technical training program related to AI development tools and does not engage with politically charged issues. It provides factual information about the course content, structure, and tools used without any apparent ideological framing or bias.

Why factuality (94): The article accurately reflects the primary source document by describing the training program's structure, objectives, and technical components such as Claude Code, Mastra, and CopilotKit. It mentions specific topics like tool calls, security practices, and integration with external systems, aligni

Why objectivity (97): The article maintains a neutral and informative tone throughout, presenting facts without bias or emotional language. It provides a balanced overview of the training program, emphasizing its educational value and practical applications without favoring any particular perspective or outcome.

How each side covered it

The same event, grouped by the political lean of the outlets covering it.

How each side covered it

Support independent, bias-aware news and unlock the social pulse, community voting, and every other Supporter feature.

Become a Supporter

Covered around the world

The same event as reported in other countries.

Covered around the world

Support independent, bias-aware news and unlock the social pulse, community voting, and every other Supporter feature.

Become a Supporter

Claims check

Key factual claims, and how many sources assert vs dispute each.

Claims check

Support independent, bias-aware news and unlock the social pulse, community voting, and every other Supporter feature.

Become a Supporter

Keep the news honest.

ObjectiveNews is reader-funded and ad-free — we show you the bias instead of hiding it. Support independent journalism for €4/month.

Become a Supporter

Related stories