Project Discovery
23h ago•sndevpostedLatest Release
Building a Simple Large Language Model from Scratch
Build a transformer-based LLM from scratch using Python and PyTorch. Master tokenization, positional encodings, multi-head self-attention, and autoregressive model training.
23h ago•sndevposted
Fine-Tuning Large Language Models (LLMs) with Python
Learn how to fine-tune an LLM (DistilGPT-2) from scratch using Python and Hugging Face Transformers. Covers tokenization, dataset preparation, Trainer API training, and inference.
sndev YouTube Channel
Watch in-depth system architectures, step-by-step implementation tutorials, and code reviews.
23h ago•sndevposted
Build a RAG Based PDF Chatbot with Fundamentals of RAG
Build a custom RAG PDF chatbot using LangChain, ChromaDB, and Google Gemini API to perform fast semantic retrieval and generate grounded, context-aware answers.
23h ago•sndevposted
End-to-End AI Agent Chatbot withModular Structure
Build a full-stack Agentic AI Chatbot using LangGraph, LangChain, Groq (LLaMA 3), OpenAI GPT-4o, Tavily Search, FastAPI, and Streamlit for real-time web-augmented reasoning.
23h ago•sndevposted
Multilingual AI Chatbot using Sarvam AI for India's Regional Languages
Build a multilingual AI chatbot using Sarvam AI and Streamlit. Supports Indian languages like Hindi, Gujarati, Bengali, and Kannada with native LLM reasoning and real-time translation.
23h ago•sndevposted
Build a Modular RAG Chatbot
Build a modular, full-stack RAG PDF chatbot from scratch using LangChain, FastAPI, Streamlit, ChromaDB, and Groq (LLaMA 3) with clean frontend-backend separation and sub-second inference.
1 day ago•sndevposted
Build an End-to-End Modular PDF RAG XML Role Validator
Automate job role validation between structured XML files and unstructured PDFs using Google Gemini, Pinecone vector search, PyMuPDF, and fuzzy string algorithms (Levenshtein Distance & Ratcliff-Obers
1 day ago•sndevposted
Build an End-To-End Modular RAG LangGraph AI Agent with Web Search
Build a full-stack AI agent with LangGraph, Groq LLaMA 3, Pinecone, and Tavily Search API. Features dynamic routing between PDF-grounded RAG and real-time web search with a FastAPI and Streamlit archi
1 day ago•sndevposted
Build a MultiModal RAG PDF Q&A Chatbot
Build a Multimodal RAG pipeline using PyMuPDF, Cohere embed-v4.0, ChromaDB, and Gemini 2.5 Flash with LangChain LCEL to query both text and visual diagrams from PDFs.
1 day ago•sndevposted
Build & Deploy an End-To-End Modular RAG-based Medical Assistant
Build a role-based medical chatbot using FastAPI, Groq LLaMA 3, Pinecone, and MongoDB Atlas. Features document-grounded RAG with strict RBAC for Doctors, Nurses, and Patients.
1 day ago•sndevposted
Build & Deploy an End-To-End Modular RAG Medical Diagnosis using LangChain, Pinecone, FastAPI
Build a secure, real-time medical report diagnosis app using FastAPI, Streamlit, LangChain, Pinecone, and Groq. Features RAG for accurate insights and RBAC for role management.
1 day ago•sndevposted
HyDE RAG Explained with Python Project | Fix Retrieval Quality in RAG
Fix RAG hallucinations using HyDE. Bridge the semantic query-document gap with FastAPI, Groq LLaMA, Gemini embeddings, and Pinecone, boosting similarity from 0.69 to 0.85.
1 day ago•sndevposted
Build & Deploy an AI Teaching Assistant with End-to-End Modular RAG System & Deployment
I Teaching Assistant is a production-ready, full-stack Generative AI application designed to streamline classroom learning through Retrieval-Augmented Generation (RAG) and automated evaluation
6 days ago•sndevposted
Traditional vs Vectorless RAG | When to use which one | Full Comparison
Discussion on Traditional RAG vs. Vectorless RAG, covering how both approaches work, their key differences, advantages and limitations, and most importantly, when to use each approach based on the use
7 days ago•sndevposted
LangGraph Fundamentals Explained with Nodes, State, Edges and Agentic AI Workflows
Fundamental concepts regarding LangGraph and concept explanation
Showing 15 projects