GenAI_Syllabus
GenAI_Syllabus
Generative AI
Bootcamp
NLP with Generative AI
This course offers a comprehensive understanding of Natural Language Processing (NLP) through
Generative AI, focusing on key concepts like language modeling and text generation. You will gain
skills in analyzing, generating, and manipulating textual data using advanced neural architectures
such as transformers. By exploring state-of-the-art techniques and tools, you'll be empowered to
develop innovative applications that harness the transformative potential of generative AI in
language interactions.
Learning Objectives
Build a strong understanding of NLP, including key concepts, techniques, and
challenges associated with generative AI.
Implement and work with state-of-the-art models such as GPT, BERT, LLAMA, and
Mistral for effective text generation and manipulation.
Utilize RAG techniques to improve contextual accuracy in text generation by
integrating external knowledge sources.
Master prompt engineering and explore the creation of multimodal applications that
combine text, images, and audio to leverage the capabilities of generative AI.
Investigate the integration of text, image, and audio data to create multimodal
applications that leverage generative AI across different media types.
Learn how to evaluate and fine-tune language models for specific tasks and gain
practical experience deploying NLP applications using frameworks like Hugging Face
and LangChain.
Module 1
Breif Overview of Classical NLP & Introduction of
GenAI
This section gives a brief overview of classical NLP and introduces GenAI. It explains the
differences between various types of neural networks like ANN, CNN, and RNN, including LSTM
and GRU. The section also covers encoder-decoder architectures that use attention mechanisms
and introduces the transformer architecture, which has changed the way we handle NLP tasks.
Generative AI is defined, highlighting its importance and core components, along with its many
applications and use cases. Finally, it discusses the ethical and social implications of Generative AI,
emphasizing the need for responsible use in technology.
Topics
Topics
Topics
Topics
Topics
Hugging Face API and Inference Learn to utilize the API for model
inference.
Model Deployment and Sharing Push models to the Hub and deploy in
Spaces.
Multimedia Models with Hugging Face Work with image, video, and audio
models.
Module 6
Overview of Major AI APIs: Setup and Key Features
This section provides an overview of major AI APIs, starting with an introduction to OpenAI APIs,
including account setup and pricing details. It covers various OpenAI models like GPT-3.5 and
GPT-4. The section also introduces Google Gemini, including its setup, pricing, and model
variations. Lastly, it discusses Anthropic Claude, detailing its key setup and available models,
offering a comprehensive understanding of these influential AI tools.
Topics
Topics
Topics
Indexing Methods for Vector Search Efficient techniques for fast vector
retrieval.
Topics
Topics
AI Agents and Model Monitoring & Deploying models using LangServe and
Deployment monitoring with LangSmith.
Module 11
Overview of LlamaIndex
This overview of LlamaIndex compares it with LangChain, highlighting its unique features. It
discusses the LlamaIndex Data Loader and Web Scraper, showcasing their roles in data handling.
The guide covers Retrieval-Augmented Generation (RAG) capabilities with LlamaIndex and
explores multimodal applications. Additionally, it introduces the concept of agents within
LlamaIndex and provides insights into Llama Hub, emphasizing its functionalities and how it
enhances the LlamaIndex ecosystem for developers and researchers.
Topics
Data Loader and Web Scraper Tools for loading data and web scraping
efficiently.
AI Agents
This guide introduces AI agents and their integration within the LangChain framework, detailing
types such as the ReAct Agent, Structured Output Agent, and Self-Ask with Search Agent. It
provides an overview of LangGraph and its components, focusing on the development of multi-
agent systems. Additionally, the guide explores Retrieval-Augmented Generation (RAG) with
LangGraph, including Corrective RAG, Agentic RAG, and Self-RAG. Finally, it highlights CrewAI
and Autogen for built-in agent capabilities, enhancing the functionality of AI applications.
Topics
Topics
Topics
Cloud Platforms for LLM Deployment Comparing cloud platforms that facilitate
efficient LLM deployment.
Module 15
Topics