AI Product Manager

Understand the importance of bespoke datasets for AI accuracy and domain-specific applications. Learn how to create and curate high-quality multimodal datasets (text, video, and audio) for AI products, combining supervised machine learning with computer vision techniques. Gain insights into the key stakeholders in AI product development, and finish with a project to develop a roadmap for an AI-powered video creation app.

Lesson 1: Supervised Machine Learning and Computer Vision

• Learn the fundamentals of supervised machine learning, including data requirements and performance evaluation.
• Explore the principles of computer vision, its historical development, and practical applications.
• Create a product proposal that integrates supervised machine learning and computer vision to address a real-world problem.
• Understand how to apply AI technologies in various products, from simple applications to complex systems like autonomous vehicles.

Course Project: Tasty Cuts: An AI-Powered  Video Creation App for Home Chefs

In this project, you will define the roadmap for an AI-powered mobile app for home chefs. You’ll break down barriers towards widespread adoption by considering the product features and prioritisation to make it easier for home chefs to create short-form cooking videos

• Define a product goal for a medical diagnostic tool
• Design an annotation job for a medical image dataset
• Consider metrics for success, how one might improve the annotation design, and design test questions for annotators

Learn the fundamentals of AI models, including how neural networks make decisions and the training process. Understand how training data impacts model performance, evaluate model results, and explore techniques like transfer learning and neural architecture search to make AI more accessible to diverse users.

Lesson 1: NLP and Conversational AI

• Learn how a neural network learns from training data
• Use test data to evaluate a trained model according to metrics like accuracy, precision and recall
• Learn how to use pre-trained models to transfer learning from one resource to another 

Course Project: Innovative Solutions with Large Language Models (LLMs)

In this project, get experience building models using automated ML, from data to results (no coding required). Build your own model using Google AutoML for a medical imaging use case. Then, implement the model with four different variants of the data to appreciate how the data affects the performance of the model.

• Build and train a model using Google’s AutoML
• Evaluate several models and decide on the best model for a given product use case