TECH 16 — Large Language Models for Business with Python
Quarter: Winter
Instructor(s): Charlie Flanagan
Date(s): Jan 29—Mar 19
Class Recording Available: Yes
Class Meeting Day: Wednesdays
Class Meeting Time: 7:00—8:50 pm (PT)
Tuition: $500
Refund Deadline: Jan 31
Unit(s): 1
Status: Registration opens Dec 2, 8:30 am (PT)
Quarter: Winter
Day: Wednesdays
Duration: 8 weeks
Time: 7:00—8:50 pm (PT)
Date(s): Jan 29—Mar 19
Unit(s): 1
Tuition: $500
Refund Deadline: Jan 31
Instructor(s): Charlie Flanagan
Recording Available: Yes
Status: Registration opens Dec 2, 8:30 am (PT)
Large language models (LLMs) help people with the everyday aspects of their lives, including writing content, increasing personal productivity, and simplifying daily tasks. By examining GPT-4, BERT, and other models, this transformative course offers an expansive and detailed understanding of LLMs and how they can be applied to create a competitive business advantage. The curriculum delves into the fundamental concepts, architectures, and training techniques required to create real-world applications, emphasizing hands-on experience using prominent platforms such as Python, LangChain, OpenAI, and Hugging Face. The course also teaches students the practical skills to create large language model applications such as automatic text generators, language translators, and models that gauge consumer sentiment toward products and brands. Additionally, students will learn the following:
- Differences between various model architectures and how to select which architecture is best suited for a particular use case
- Techniques for efficient training and fine-tuning of models
- Selecting and interpreting metrics that communicate how accurately the model makes predictions on new data it wasn’t trained on
Students are expected to have a basic understanding of Python and machine learning. Prior exposure to natural language processing would be beneficial but is not required.
CHARLIE FLANAGAN
Head of Applied AI, Balyasny Asset Management
Charlie Flanagan is the head of data science at Balyasny Asset Management, a large multistrategy hedge fund. Earlier, he worked for Google, where he was the data science lead for Google Duplex. He received an MS in software engineering from Harvard and an MBA from Columbia. Textbooks for this course:
There are no required textbooks; however, some fee-based online readings may be assigned.