In this module, you will build a complete retrieval-augmented generation (RAG) pipeline from scratch, learning how to preprocess data, perform vector indexing, and integrate retrieval and generation using LangChain and Chroma DB. You'll gain hands-on experience with essential libraries, understand the flow of data through the pipeline, and execute queries to see RAG in action.

This module explores real-world implementations of retrieval-augmented generation (RAG) in areas such as automated reporting, e-commerce, knowledge management, and innovation scouting. Learners will discover how RAG enhances data analysis, personalizes content, and improves the utility of knowledge bases. Practical exercises will guide you in integrating sources into RAG pipelines for robust, transparent AI solutions.

This module explores the essential building blocks of retrieval-augmented generation (RAG) systems, including indexing, retrieval, prompt engineering, LLM integration, and user interface design. Learners will gain practical insights into how these components interact to create effective RAG applications. By the end, you'll understand both the technical and user-facing aspects necessary for building robust RAG solutions.

This module explores the unique security risks associated with retrieval-augmented generation (RAG) applications, including challenges posed by large language models and external data sources. Learners will investigate common vulnerabilities, such as hallucinations and sensitive information disclosure, and gain hands-on experience with red teaming and defensive strategies. Practical coding labs provide opportunities to secure API keys and implement protective measures against attacks.

This module introduces the fundamentals of building applications with retrieval-augmented generation (RAG) and demonstrates how to leverage Gradio for creating user-friendly interfaces. Learners will explore the advantages of Gradio, its integration with popular machine learning frameworks, and practical steps for interfacing with RAG models.

This module explores how vectors and vector stores underpin retrieval-augmented generation (RAG) systems, delving into vector representations, embedding models, and the practical considerations for choosing and using vector stores. Learners will gain insights into the impact of vector dimensions, semantic algorithms, and performance factors in real-world RAG applications.

This module explores the principles and techniques behind similarity searching using vector representations. Learners will examine semantic versus keyword search, distance metrics like Euclidean distance, and various search paradigms including dense, sparse, and hybrid approaches. Practical labs and real-world tools such as Pinecone and LangChain will help solidify understanding of indexing, search algorithms, and vector search services.

This module guides learners through the quantitative evaluation of retrieval-augmented generation (RAG) systems using standardized frameworks and visualization tools. You will implement the ragas library to generate synthetic ground truth data, assess retrieval and generation metrics, and explore additional evaluation techniques to optimize RAG pipelines.

This module explores the essential building blocks of retrieval-augmented generation (RAG) systems within LangChain, focusing on vector stores, retrievers, and large language models (LLMs). Learners will gain hands-on experience with popular retriever options and understand how these components interact to enable effective information retrieval and generation.

This module explores advanced techniques for enhancing retrieval-augmented generation (RAG) workflows using LangChain. Learners will dive into practical tools such as text splitters and output parsers, gaining hands-on experience with LangChain Expression Language (LCEL) to optimize document processing and result formatting.

This module explores how to enhance retrieval-augmented generation (RAG) pipelines by integrating AI agents using LangGraph. Learners will discover how graph theory concepts, agent state management, and decision-making nodes can be leveraged to build more dynamic and intelligent workflows. Practical coding exercises guide you through implementing and customizing agentic RAG systems.

This module explores effective prompt engineering techniques to enhance retrieval-augmented generation (RAG) systems. Learners will discover strategies for designing, adapting, and optimizing prompts for various large language models, and practice applying these concepts to tasks such as summarization, data extraction, transformation, and expansion.

This module delves into advanced strategies for enhancing retrieval-augmented generation (RAG) systems, including re-ranking, query decomposition, and multi-modal RAG techniques. Learners will gain hands-on experience with code labs and explore methods for integrating text and image data to improve GenAI applications.