GenAI Projects Collection

A comprehensive collection of AI and machine learning projects, experiments, and learning materials focused on large language models, agentic workflows, and advanced NLP techniques.

📁 Project Structure

Core Agent & Workflow Projects

• DL_Intro_to_Agentic_Workflows - Foundational agentic workflows including customer service, email assistance, and market research implementations
• DL_Agents with Crew AI - Multi-level agent projects using Crew AI framework for collaboration, research, event planning, and financial analysis
• DL_Agentic_graph - Knowledge graph construction for agentic systems with schema proposal and intent recognition
• DL_deep_agents - Advanced deep learning agent implementations
• DL_MultiAgent - Multi-agent system architectures and patterns

LLM & Foundation Model Projects

• DL_HuggingFace_OpenSourceModels - Comprehensive HuggingFace model applications including NLP, translation, embeddings, speech, and vision tasks
• DL_Llama4 - Llama model implementations and fine-tuning examples
• DL_LLM-post-training - Post-training techniques and optimization strategies
• DL_openai o1 - OpenAI o1 model exploration and usage
• SFT_Finetune_Gemma - Supervised fine-tuning for Gemma models
• gemma3_finetune_unsloth - Efficient Gemma 3 fine-tuning with unsloth

Data & Retrieval Systems

• DL_RAG - Retrieval-Augmented Generation implementations
• DL_LangChain_data - LangChain tutorials covering document loading, splitting, and embeddings
• DL_Vector_embeddings - Vector embedding techniques and applications
• RAG_pipe - RAG pipeline implementations
• RAG_rerank - Reranking strategies for RAG systems
• rl_recommendar - Reinforcement learning for recommendation systems

Specialized Techniques & Frameworks

• DL_DSPy - DSPy framework for prompt optimization and program synthesis
• DL_Pydantic - Pydantic for data validation and structured outputs
• guardrails - Safety and guardrail implementations for LLM outputs
• GRPO - Group Relative Policy Optimization experiments
• dspy_gepa_optimization-main - DSPy optimization techniques

Deployment & Miscellaneous

• azure_agent - Azure-based agent implementations
• function_gemma_hf - Function calling with Gemma models
• kaggle_notebooks - Kaggle competition notebooks and solutions
• langgraph_agents - LangGraph-based agent implementations

🚀 Getting Started

Each project folder contains:

• Jupyter Notebooks (.ipynb) - Interactive learning materials and implementations
• Python Scripts - Reusable utilities and helper functions
• requirements.txt - Project-specific dependencies

Prerequisites

• Python 3.8+
• Jupyter Notebook or JupyterLab
• API keys for relevant services (OpenAI, HuggingFace, etc.)

Installation

1. Clone or navigate to the repository
2. Create a virtual environment (recommended):

   python -m venv venv
   source venv/bin/activate

3. Install dependencies for your project:

   pip install -r path/to/requirements.txt

📚 Learning Path (Suggested)

1. Start with Fundamentals

   • DL_Intro_to_Agentic_Workflows
   • DL_LangChain_data
   • DL_Vector_embeddings

2. Explore Foundation Models

   • DL_HuggingFace_OpenSourceModels
   • DL_Llama4

3. Advanced Techniques

   • DL_Agents with Crew AI
   • DL_RAG
   • DL_DSPy

4. Fine-tuning & Optimization

   • SFT_Finetune_Gemma
   • DL_LLM-post-training
   • DL_Pydantic

🔧 Common Workflows

Running a Jupyter Notebook

jupyter notebook path/to/notebook.ipynb

Installing Project Dependencies

cd path/to/project
pip install -r requirements.txt

Running Python Scripts

python script_name.py

📝 Notes

• Each project may have its own specific setup requirements
• Some projects may require API keys or external credentials
• Check individual project folders for specific README files or documentation
• Keep requirements.txt files updated when adding new dependencies

🎯 Key Technologies

• LLMs: OpenAI, HuggingFace, Llama
• Frameworks: LangChain, LangGraph, Crew AI, DSPy
• Databases: Neo4j (for knowledge graphs)
• Techniques: RAG, Fine-tuning, Agentic Workflows, Knowledge Graphs

---

Last Updated: February 2026

For detailed information about specific projects, navigate to their respective folders and check for project-specific documentation.