Coursera co-founder, Andrew Ng, is launching a new course on Agentic AI at DeepLearning.AI

Andrew Ng, one of the pioneers in the field of AI, co-founder of Google Brain and Coursera, is launching a new course on Agentic AI, focused on developing autonomous AI systems capable of performing complex, multi-step tasks.

Unlike the traditional approach, where AI simply responds to individual queries, agentic workflows enable models to plan and execute a sequence of actions, improving outcomes through reflection and interaction with tools.

The course aims to provide participants with fundamental knowledge and practical skills for building complete agentic applications, from workflow design to deployment and performance evaluation. The course helps break down business processes into multi-step AI processes, identifying opportunities for automation and integration with external services.

What participants will learn: 

— reflection: AI systems that independently evaluate their own performance, correct errors, and improve task execution quality

— tool use: connecting AI to databases, APIs, and external services to perform actions, rather than just generating text

— planning: breaking down complex tasks into step-by-step processes that AI can execute and adapt to changing conditions

— multi-agent: coordinating multiple specialized AI systems to solve complex problems and optimize workflows.

Practical Focus

Participants will work with Python, building agentic patterns from scratch and gradually integrating them with frameworks and external tools. Special emphasis is placed on real-world implementation: participants will learn to develop testing frameworks, conduct error analysis, and optimize systems for production deployment. The course includes a capstone project—building a fully functional research agent capable of collecting data, analyzing information, and generating reports autonomously.

Target Audience:

— developers who want to create autonomous systems for performing complex tasks

— professionals with intermediate Python skills, ready to complete practical assignments

— specialists familiar with large language models and APIs, looking to enhance their practical skills.

The program consists of five modules featuring self-paced learning and practical assignments for building agentic AI systems. Upon completion, you will receive a certificate confirming your skills in developing such workflows.

The course is available on the DeepLearning.AI platform and will be beneficial for developers and AI specialists who want to improve their skills and learn how to apply modern methods for building autonomous intelligent systems.