Engineering Autonomous AI Agents Master Class

Speakers

• Thomas Martinsen

About this session

Most AI solutions today assist humans. The next generation of systems will execute work independently. This masterclass provides a deep dive into how to engineer autonomous AI agents that can reason, plan, and take action across real business processes. In this hands-on masterclass, you will learn how to build AI agents capable of reasoning, planning, and acting autonomously across complex business processes. Through demos and practical exercises, you will create agents that can discover information, gather and evaluate data from multiple sources, build knowledge, and dynamically determine the best path toward achieving a goal. Unlike traditional automation, autonomous agents are not limited to predefined workflows. They can adapt to changing conditions, identify knowledge gaps, seek out the information they need, and continuously refine their approach while executing tasks on behalf of users and organizations. By the end of the masterclass, you will have built and operated autonomous AI agents yourself and gained a practical understanding of how to create systems that move beyond assisting people to actively performing work. The future of AI is not smarter chat interfaces—it is autonomous digital workers capable of turning objectives into outcomes. This is not a low-code agent builder session. The masterclass uses a code-first engineering stack for creating autonomous AI systems, covering large language models, agent orchestration, tool use, retrieval, memory, APIs, data integrations, evaluation, and observability. Participants will work across modern AI engineering environments including VS Code, GitHub Copilot, Claude Code, and Codex—using both IDE-based and CLI-based workflows. The focus is on understanding the technical building blocks required to move from AI demos to autonomous agents that can operate reliably in real enterprise contexts.