Course Topics
This in-person, hands-on course introduces participants to the dynamic world of Large Language Models (LLMs) through a practical lens, focusing on how to use and implement these technologies effectively.
Starting with an overview of the fundamental concepts of language models, such as preprocessing and basic neural network architecture and transformer models, the curriculum quickly moves into practical applications, demonstrating how LLMs can be leveraged in diverse real-world scenarios, from AI-driven customer service to content creation.
Click here to view the course schedule.
Course Format
Through lectures, guided code-based tutorials, and interaction with industry contributors, participants will gain first-hand experience interacting with, modifying, and using models through popular APIs including from Hugging Face and OpenAI, along with advanced techniques like Retrieval-Augmented Generation (RAG), vector databases, agentic frameworks, and fine-tuning using LoRA.
Course Outcomes
By the end of this course, learners will be well-prepared to craft effective prompts to apply these technologies creatively and effectively in their careers or advanced studies in AI. Learners will also have joined a community of peers leading advances in applied AI.
Learners who attend the complete course will receive a participation certificate from Harvard John A. Paulson School of Engineering and Applied Sciences.
Learning Objectives
- Develop AI-driven customer service applications
- Improve content creation workflows using LLMs
- Implement advanced LLM techniques like RAG and fine-tuning
- Integrate LLMs with other systems using APIs
- Build and deploy agentic LLM frameworks
Course Prerequisites
- Programming Skills
- Proficiency in Python is essential, as the workshop includes extensive coding exercises and API implementations
- Familiarity with frameworks like PyTorch or TensorFlow
- Core Knowledge in AI and Machine Learning
- Understanding of basic concepts in AI and machine learning, including supervised learning, unsupervised learning, and model evaluation metrics
- Familiarity with natural language processing (NLP) principles
- Experience with Neural Networks
- Prior exposure to neural network architectures, especially transformer models such as BERT or GPT, is highly recommended
- API Integration Experience
- Basic experience using APIs in Python to interact with external services will be helpful for the hands-on components
- Optional Recommended Skills
- Knowledge of Git and GitHub for version control
- Familiarity with the command line for setting up environments and executing scripts
Who Should Participate
Professionals in AI, developers interested in practical AI applications, and advanced undergraduate and graduate students can benefit from taking this course. This course is designed for learners from all types of organizations who want to increase their individual effectiveness or who have functional responsibilities, including:
- Machine Learning Engineers
- Data Scientists
- Data Engineers
- AI/ML Research Scientists
- Natural Language Processing (NLP) Engineers
- AI/ML Product Designers or Managers
- Solutions Architects
- Data Analysts
- AI/ML Technical Program Managers
- Content Strategists
- Automation Engineers
- Innovation Strategists
- Chief Technology Officers
Harvard University welcomes individuals with disabilities to participate in its programs and activities. If you would like to request accommodations or have questions about the physical access provided, please contact ProfEd@seas.harvard.edu in advance of your participation. Requests for American Sign Language interpreters and/or CART providers should be made at least two weeks in advance, if possible. Please note that the University will make every effort to secure services, but that services are subject to availability.
For more information, contact SEAS Professional Education at ProfEd@seas.harvard.edu.