A comprehensive repository containing tutorials, projects, and notebooks for Large Language Models (LLMs), Generative AI, and Transformer architectures.
This repository is designed for engineers, data scientists, and developers looking to master the full lifecycle of LLM development—from fundamental concepts and fine-tuning to advanced deployment and monitoring.
This repository is organized into distinct folders covering major components of the LLM ecosystem.
| Directory | Focus Area | Description |
|---|---|---|
| LLMs_from_Scratch | Foundations | Deep dives into the inner workings of neural networks and transformer architecture. |
| HandsOnLLMs | Fine-Tuning Techniques | Practical notebooks on modern fine-tuning methods like LoRA, PEFT, and Reinforcement Learning techniques such as PPO and DPO (Direct Preference Optimization). |
| ChromaDB_semantic_search | Vector Databases & RAG | Implementations of semantic search and Retrieval-Augmented Generation (RAG) using ChromaDB. |
| mcp-rag-system | Advanced RAG | Contains a specialized RAG application setup (likely involving Milvus/MCP server). |
| LLMs_deployment | Cloud & API Deployment | Examples for deploying LLMs, including a Perplexity-style clone application demo. |
| Fastapi_aws_deploy | Production Deployment | Deployment guides for LLMs (e.g., Llama) using FastAPI and AWS (potentially via Cerebrium). |
| OpenAI_streamlit_app | Front-end Applications | A complete Streamlit application showcasing an OpenAI chatbot implementation. |
| gemma_streamlit_app | Model-Specific Apps | Local deployment of the Gemma model using a Streamlit front-end. |
| Email_drafter_agent_FastAPI | Agentic Workflow | Building a practical AI agent (e.g., an email drafter) exposed via a FastAPI service. |
| NVIDIA_CUDA_BASICS | GPU Acceleration | Tutorials and custom kernels for understanding and optimizing operations with NVIDIA CUDA. |
| RAPIDS_Data_Science | GPU Data Science | Examples using the RAPIDS suite for accelerating data science workflows with CUDA/Python. |
| DeepLearningFiles | General Deep Learning | Files for broader DL concepts, such as multi-output model training using Keras. |
| grafana-prometheus implementation | Monitoring | Setup for observing and monitoring LLM services using Grafana and Prometheus. |
This project extensively uses the following tools and libraries:
- Models: Llama, Gemma, GPT (via OpenAI API)
- Frameworks: Hugging Face Transformers, PEFT, LoRA, FastAPI, Streamlit
- Vector DBs: ChromaDB
- GPU Tools: NVIDIA CUDA, RAPIDS
- MLOps/Monitoring: Grafana, Prometheus
- Cloud: AWS
In addition to the practical notebooks, this repository includes supporting files to guide your learning journey:
GenAI-blogs.md: Curated list of informative blogs and articles on Generative AI.LLM-GenAI-Courses.md: A collection of suggested courses and learning resources.- Fine-Tuning Notebooks: Dedicated notebooks focusing on advanced GPT and open-source LLM fine-tuning strategies.
This repository is licensed under the Apache-2.0 License—see the LICENSE file for details.
If you find this repository helpful, please consider giving it a star ⭐