AI Reasoning Models in Practice: Building an AI-Powered Coach

AI reasoning models are changing how we interact with AI, providing more structured, logical, and multistep responses compared to traditional GPT models. In this project-based course, learn when and how to use reasoning models as you create an AI-powered personal coach. Instructor Kesha Williams—a leader in enterprise architecture and AI strategy and governance—provides hands-on challenges that allow you to apply reasoning models to decision-making, code understanding, and image-based reasoning while optimizing model efficiency. Whether you’re a developer or data scientist looking to integrate advanced reasoning models into practical solutions—or a technical decision-maker interested in the capabilities and applications of AI reasoning models—this course can help you integrate this exciting technology into real-world AI workflows.

This course is integrated with GitHub Codespaces, an instant cloud developer environment that offers all the functionality of your favorite IDE without the need for any local machine setup. With GitHub Codespaces, you can get hands-on practice from any machine, at any time—all while using a tool that you’ll likely encounter in the workplace.

Learning objectives

• Differentiate AI reasoning models from traditional GPT models and compare leading reasoning models such as OpenAI o1 and o3, DeepSeek-R1, Gemini 2.0 Flash, and Grok 3.
• Apply best practices for prompting reasoning models, including structuring prompts effectively and using meta-prompting techniques to refine AI responses.
• Utilize AI reasoning models for real-world tasks, such as decision-making, code understanding, and image-based reasoning, selecting the best model for each scenario.
• Manage AI reasoning models efficiently by optimizing reasoning tokens, controlling context windows, and ensuring accuracy and fairness in model outputs.
• Implement hands-on techniques to integrate reasoning models into practical AI workflows.