Agentic AI Fundamentals: Build Autonomous AI Systems

Design AI systems that can think and act independently

AI is no longer just responding - it’s starting to act.

A new generation of systems, known as agentic AI, can plan tasks, use tools, and make decisions in ways that go far beyond traditional chatbots. These systems don’t just generate answers, they work towards goals.

On this course, you’ll step into this emerging space and learn how AI agents are built, how they operate, and what makes them fundamentally different from other AI systems. Rather than focusing only on theory, the course centres on experimentation and creation, giving you the opportunity to build and explore agentic systems in practice.

See how AI moves from response to action

Explore the shift from prompt-based AI to systems that can reason, plan, and act.

You’ll unpack the core architecture behind AI agents, including the agent loop, decision-making processes, and how large language models interact with tools.

This introduces a new way of thinking about AI - not as a tool, but as an active system.

Build agents that interact with the real world

Move quickly into hands-on development, creating your own AI agents from scratch.

You’ll experiment with Python, APIs, and frameworks such as LangChain to design systems that can search, retrieve information, and perform tasks.

By integrating tools and functions, your agents begin to operate beyond the limits of static models.

Explore the future of autonomous AI systems

Extend your understanding by working with multi-agent systems and exploring how agents collaborate.

You’ll also examine emerging standards, real-world applications, and key challenges around safety, control, and ethics.

This wider perspective helps position agentic AI within the rapidly evolving landscape of intelligent systems.

This course is for developers, data professionals, product managers, and tech leaders curious about the next generation of AI. It’s ideal for those exploring agentic AI or looking to build systems that go beyond traditional chatbots.

Learners need a computer (Windows, macOS, or Linux) with at least 8 GB RAM and a stable internet connection. Python 3.10 or later should be installed, along with a code editor such as VS Code. No paid software or specialised hardware is required.

This course is for developers, data professionals, product managers, and tech leaders curious about the next generation of AI. It’s ideal for those exploring agentic AI or looking to build systems that go beyond traditional chatbots.

• Explain the core principles of agentic AI, including how AI agents use reasoning, tools, prompts, memory and observations to work towards a goal.
• Build a scratch ReAct agent in Python using Groq, including tool registration, action parsing, scratchpad updates and final answer generation.
• Extend an agent workflow by adding tool-based capabilities, such as web search, to support external information retrieval and more grounded responses.
• Develop LangChain and Gemini-based agent workflows, including a single-agent Meeting Notes Assistant and a multi-agent Researcher–Writer–Reviewer workflow.
• Evaluate agentic AI systems by comparing frameworks, multi-agent patterns and responsible AI guardrails for reliability, safety, scalability and production readiness.