Courses

Like children, large language models don't always behave as expected. At times they can be unhelpful, nonsensical, and unintentionally offensive. And, as with children, the best training doesn't guarantee their success. Their development requires the abilities of a scientist, including curiosity, discipline, and problem-solving skills. But their unpredictable nature also requires the qualities of a parent, including patience, positive reinforcement, and open-mindedness. It's a complex and challenging endeavor, but when done right, the results can be wildly rewarding and widely beneficial.

The production process of large language models is particularly difficult and requires an array of skills and knowledge, including machine learning algorithms, data engineering, software engineering, and ethical considerations. This course bridges the gap between classroom learning and industry practices through hands-on experience and real-life examples, including generating human-like text for creating more realistic video game characters and advancing personalized medicine by analyzing a patient's medical history and genetic data to identify which potential treatments are more likely to be effective. By learning how to put machine learning models into production using industry practices, you'll learn not only sought-after skills in machine learning model design, but also how to formulate holistic ML strategies when tackling complex real-world problems.

This course will cover the following:

• Comprehensive understanding of large language models
• Practical applications of large language models in the real world
• Challenges that arise when working with messy data
• Flexible approaches for selecting the best model for a particular problem
• Real-life examples of putting machine learning models into production
• Developing skills in machine learning model design and implementation
• Strategies for overcoming common challenges in machine learning system design