Gen AI - LLM RAG Two in One - LangChain + LlamaIndex

Explore how LangChain acts as a flexible framework that connects diverse large language models to your data sources and enables retrieval augmented generation through chains and memory.

Set up a smooth LangChain project by creating a Python virtual environment with Pipenv, configuring a dot env for the OpenAI API key, and using a VS Code runner.

Set up your Python virtual environment and install essential libraries for Gen AI LLM RAG work, including pandas, python-dotenv, OpenAI, LangChain, and more, using pipenv or pip install.

Learn to use Google Gemini LLM as a free alternative to OpenAI in rag applications, including API key setup, dot env configuration, and LangChain integration.

Explore building your first LangChain program with a two part chain comprising a prompt template and an LM call, tested with OpenAI GPT turbo and Gemini LM.

Build a rag application that translates natural language into well-formed SQL queries. Connect to a MySQL analytics database and run queries like employee counts and top shipping charges.

Build a CV summary generator and RAG app with pdf/docx uploads, OpenAI GPT API and LangChain, exposed via a Streamlit UI.

Create an invoice extract rag application that ingests PDFs, extracts structured data with GPT via LangChain, outputs a dataframe and downloadable CSV via Streamlit UI.

Build a Rag application with LangChain for conversational HR policy queries, employing memory, OpenAI embeddings, and a vector database to retrieve relevant information from web page data.

Analyze employee attrition data from a csv using a Streamlit UI, pandas data frames, and a zero-shot react description agent in LangChain to perform multi-step analysis and insights.

Explore how lama index, a context augmented LLM framework, enables retrieval augmented generation with vector stores, data connectors, and query engines to build AI applications.

Set up the llamaindex development environment by creating a virtual environment, configuring dot env with API keys, and validating via sample OpenAI code in VS Code.

Ingest three PDFs with a data connector, create a vector store index from text embeddings, and enable querying via a retrieval object and query engine. Apply post-processing to compact responses.

Build a rag app using chroma db as a persistent vector store to index 51 pages from three mental health pdfs, comparing OpenAI and Gemini outputs with adjustable context window.

learn how llama index enables rag for sql databases by querying a walmart table using three mechanisms: sql query engine, sql table retrieval query engine, and natural language sql retriever.

Explore LlamaIndex query pipelines and how prompts, retrieval, reranking, and response synthesis power rag applications with observability and modular components.

Learn to build a simple sequential query pipeline in LlamaIndex using a prompt template and OpenAI model, loading the API key and executing the pipeline to fetch movie details.

Explore a complex LlamaIndex DAG pipeline that combines a vector store, a retriever, a reranker, and a tree summarizer to answer content exposure risk questions from mental health documents.

Set up a data frame pipeline with llama index to run pandas-based queries, parse outputs, and synthesize responses. Explore obesity dataset attributes and analyses like correlation and gender distribution.

Explore how Lama index agents use tools and a query engine to enable retrieval augmented generation, accessing external data from sql tables to vector stores, augmenting prompts.

Build a calculator using react agents and tools, defining Python functions like add, multiply, sine, cosine, and tangent, and solve multi-step problems via LangChain and LlamaIndex prompts.

Create a document agent with dynamically built tools that query three pdfs using per-document vector indexes, enabling the agent to route queries to the right source.

Create a code checker RAG app with a Streamlit user interface, demonstrating a simple RAG workflow that checks code across languages and returns feedback.

Build and deploy a production-ready LangChain app by creating a FastAPI backend, integrating a Streamlit UI, using in-memory vector storage, text splitting, and OpenAI embeddings.