Explore large language models and transformer architecture, their applications from chatbots to document question answering with RAG, and the tradeoffs of proprietary versus open source models.

Demystifies transformer architecture, showing how self-attention powers modern LLMs and how encoders and decoders convert text to embeddings and generate output.

Learn to call OpenAI LLMs in Python, access proprietary and open-source models, and integrate with frameworks like long chain and llama index, whether in the cloud or locally.

Explore grok, an open source llm platform, and learn to set up its Python interface, obtain an API key, and run your first request with Llama or Deep Seek.

learn to run open-source language models locally with Ollama, installing, loading models like gamma two billion, and interacting via Python and Jupyter Lab for a developer-friendly local LLM workflow.

Learn to access multiple llms from OpenAI, Gemini, and Grok through a single LangChain interface. Build prompts, memory, data connections, and pipelines while easily switching providers for real-world AI apps.

Explore few-shot prompting with context and examples to guide language models. Compare zero-shot and few-shot strategies for consistent formatting and classification outputs.

Build a fully functional GenAI chatbot with a streamlit user interface and preserve chat history in the browser via a simple session state.

Learn how retrieval augmented generation overcomes llm limitations by linking to external knowledge sources, using document ingestion, embeddings, and a vector database to provide reliable, up-to-date, grounded answers.

Explore AI agents as autonomous systems that perceive, reason, plan, and act to achieve user-defined goals using tools, memory, and APIs, with real-world applications and popular frameworks.

Build an AI agent with pedantic AI by defining inputs and outputs, registering custom tools, and executing a weather forecast tool to fetch current conditions.

Build a weather agent with Microsoft's Autogen to show how multi-agent collaboration, conversation driven workflows, and easy tool integration power real-time weather queries via external APIs.

Master multi-agent systems with crew AI, coordinating a stock research agent and a trader agent to use live market data and decide buy, hold, or sell.

Deploy open source LLMs locally and in production using Docker and Ollama. Move to cloud with EC2, RunPod, and VLM, and expose models via a FastAPI REST API.

Deploy Ollama llms on AWS EC2 with Docker to run llama models in a GPU-enabled cloud. Expose port 11434 and connect from your local machine to access the model.

Wrap locally running Ollama models with a FastAPI server to expose chat endpoints for front-end apps, using env configuration, pydantic schemas, and dockerizing with Docker Compose.

Build an MCP server in Python to wrap an existing weather tool, learn MCP host and client roles, and connect it to Cloud Desktop using stdio transport.

Build a private, customizable ChatGPT clone with a streamlit UI, local llm hosting, and mongodb-backed chat history, deployable via docker or aws ec2 for secure, private use.

Set up the ConvoPro project by creating a GitHub repository and configuring a git workflow. Link PyCharm and install a virtual environment, requirements, env templates, MongoDB, llama models.

Learn to implement a self-contained generative AI chat app with ConvoPro by building modular components—config, MongoDB-backed DB, LM factory, and Streamlit UI—loading llama models locally and generating titles for conversations.

Deploy ConvoPro on an EC2 instance to share a public chat interface. Configure GPU-accelerated models, MongoDB, and a Streamlit app via Docker and GitHub deployment.

Develop a rag-powered ai study assistant with subject and chapter selection, simple-language explanations, video references, chat history, deployable on Streamlit with Grok LLM, Chroma DB, LangChain, and AWS EC2.

Set up study pal environment by creating a GitHub repo, cloning it, configuring a Python virtual environment, and installing libraries from requirements.txt for ingestion and chat with LangChain and Chroma.

Build a AstraRAG grounded chatbot with backend and frontend, using fast API endpoints and Streamlit UI, enabling document ingestion, knowledge grounding, and explainability with sources, tools used, and rationale.

Deploy the AstraRAG chatbot with Docker by building a Docker image from a Dockerfile, running containers for the API and Streamlit frontend, and deploying to EC2 via GitHub steps.

Conclude your journey in generative AI by mastering chatbots, AI agents, deployment, and MCP, linked to capstone projects, with ongoing appendix modules and future tool updates to stay prepared.