Introduction to Building LLM powered APPs

In today’s rapidly evolving technology ecosystem, Large Language Models (LLMs) stand as a significant advancement in the realm of AI. The ability of LLMs to process vast amounts of information, understand context, generate human-like text, and interface with a variety of applications has redefined the boundaries of AI applications. This course invites you to explore the depth of LLM powered applications and become adept at leveraging these models for diverse and innovative solutions. With tools like LangChain simplifying the process of building these applications, now is the perfect time to immerse yourself in this dynamic and emerging field.

With the knowledge and skills imparted through this course, participants will be well-equipped to venture into the world of LLM powered apps, bringing innovation and expertise to the fore. Join Sawy on this exciting journey to redefine the limits of AI applications.

Prerequisites

• Curiosity
• Basic arithmetic skills - Brackets, division, multiplication, addition, subtraction
• Ability to operate a computer, keyboard and mouse
• Ability to use a web browser to access and use the internet
• Ability to install software on your computer

Hardware and Software Requirements

• Physical operational computer (not in virtualization) – Fedora 34 or greater OR PopOS/Ubuntu 20.04 or greater, OR Windows 10 or greater, OR MacOS 10 or greater
• 16 GB RAM
• Broadband internet connection > 5 MBPS
• 50 GB free hard disk space. SSD Drive recommended

Learning Objective

• Gain a foundational understanding of LangChain and its components.
• Explore the OpenAI API and its features.
• Dive deep into ChatGPT and its capabilities.
• Learn how LangChain assimilates with external resources and empowers LLMs.
• Understand the advantages and flexibility LangChain brings to LLMs.
• Familiarize oneself with the key applications of LangChain and LLMs.
• Explore Pinecone, a vector database, and understand its role in LangChain embeddings.
• Embark on project work to apply the theoretical knowledge in real-world scenarios.

Learning Outcome

Upon successful completion of this course, participants will be able to:

• Effectively utilize the OpenAI API to harness the power of ChatGPT and other LLMs.
• Build web applications that utilize LLMs using LangChain.
• Modify and customize components crucial for language models with the modularity that LangChain offers.
• Create ‘chains’ or sequences of linked actions tailored to specific needs using LangChain.
• Implement LLMs in applications ranging from QnA bots to summarization and querying data tables.
• Interface LLMs with external systems and APIs such as Wikipedia, Youtube and Google.
• Set up chatbot systems and agent-based systems powered by LLMs.
• Demonstrate hands-on proficiency through diverse project work.

Course Outline

1.Introduction to LLMs & OpenAI API

• What are LLMs?
• Secrets of Generative AI using ChatGPT.
• Overview of OpenAI and its API.

2. Dive into LangChain Basics

• LangChain’s structure for crafting applications.
• Incorporating external resources with LLMs.
• Empowering LLMs for decision-making and interaction.

3. The Vector Database

• Basics of Pinecone, Chroma and FAISS.
• Role of embeddings in retrieval.

4. Advantages & Key Applications of LangChain

• Modularity, sequences, and agility.
• Summarization, extraction, evaluation, and more.

5. Practical Insights and Hands-on Sessions

• Working with default models like davinci-003 and GPT-3.5-turbo and GPT-4.
• Dive into non-trivial example codes and live hands-on guided projects.
• Building web apps using Streamlit.

6. Project Work Student will undertake one of the following use-cases as projects:

• Summarization projects.
• Advanced retrieval with LangChain.
• Clean and standardize data.
• Work with call or video transcripts.
• QnA System on custom data.

Fineprint

• The topics presented are tentative and we reserve the right to add or remove a topic to update or improve the bootcamp, or for a technical or time reasons.
• † taxes extra.