Developing Application with Large Language Model
About This Course
This intensive, hands-on workshop will provide a comprehensive overview of developing applications with large language models (LLMs). Over the course of three days, participants will learn the fundamental concepts of transformer models and how to engineer effective prompts to direct LLMs. We will explore the full lifecycle of a generative Artificial Intelligence (AI) project from ideation to deployment. Through practical sessions, attendees will gain expertise in training, evaluating and deploying LLMs for custom use cases. The workshop will also cover core components of LangChain and equip participants with the skills to set up the environment, manage memory and embeddings, conduct Quality Assurance (QA) over documents, integrate tools and callbacks, deploy local models, utilize open-source LLMs like llama 2, and ultimately deploy LLM applications. Participants will leave equipped to build their own solutions leveraging the power of modern language processing.
What You'll Learn
• Understand the fundamental concepts and architecture of Transformer models, which are foundational to modern language processing
• Acquire the ability to engineer effective prompts that can direct generative Artificial Intelligence (AI) models to produce specific desired outcomes
• Explore the full lifecycle of a generative AI project, encompassing stages from conceptualization to deployment
• Develop expertise in the training and evaluation of large language models (LLMs) for tailored use cases.
• Learn the essential components of LangChain
• Gain practical experience in the setup and installation of LangChain, preparing the environment for language processing tasks
• Implement and utilize memory strategies within LangChain to maintain contextual awareness across conversational interactions
• Use document loaders and transformers to handle and convert textual data into a format suitable for processing by LLMs
• Master the creation of text embeddings, their management in vector databases, and their role in information retrieval
• Conduct hands-on Question & Answer sessions over documents using LLMs and assess the applications’ performance for accuracy and relevance
• Dive into the advanced functionalities of LangChain, learning to integrate agents, tools, callbacks, logging, and providers for robust application development
• Learn to integrate LangChain with local LLMs for building and deploying language models directly on personal or on-premise hardware
• Explore the utilization of open-source LLMs like llama 2 within LangChain, understanding the benefits and limitations of open-source models in the development of language applications
• Execute the deployment of LLM applications
Entry Requirements
Proficiency in English &
Preferred:
- Basic Python programming
- Experience working with JSON formats
- Comfort using command line interfaces
- Understanding of REST APIs