Pinned Loading

1. LangGraphwise LangGraphwise Public

   LangGraphwise is an educational curriculum for building AI agents with LangGraph, MCP, and Ollama. 🎓 It provides step-by-step progression from beginner to advanced agent development with local-firs…

   Jupyter Notebook 2

2. LoRA-From-Scratch LoRA-From-Scratch Public

   LoRA-From-Scratch is a pure PyTorch LoRA implementation built from scratch without PEFT dependencies. 📚 Designed for educational deep-dives, it runs on free-tier Tesla T4 GPUs while teaching LoRA i…

   Jupyter Notebook 3

3. nanoGPT nanoGPT Public

   Forked from karpathy/nanoGPT

   The simplest, fastest repository for training/finetuning medium-sized GPTs.

   Python

4. RAGOps-Tracker RAGOps-Tracker Public

   RAGOps Tracker is a RAG evaluation pipeline integrating MLflow, DVC, and RAGAS for comprehensive experiment tracking. 📊 It measures answer relevance, correctness, faithfulness, and harmfulness metr…

   Python

5. RLHF-Trainer RLHF-Trainer Public

   RLHF Trainer implements the complete RLHF pipeline for consumer GPUs: supervised fine-tuning → reward modeling → PPO training. 🎯 Running on 2080Ti 12GB, it brings ChatGPT-style alignment to your lo…

   Python

6. Agentrix Agentrix Public

   Agentrix is a LangGraph-powered multi-agent system using supervisor patterns for task delegation. ⚡ It includes LangSmith evaluation integration, human-in-the-loop verification, and ready-to-use cu…

   Jupyter Notebook 1