
A production-ready RAG system that continuously ingests web content asynchronously. Features context-aware querying via Gemini API and high-performance vector search.
FastAPI
Celery
Redis
QdrantDB
Next.js
PostgreSQL
Docker
Nginx

An automated contract analysis tool using Tesseract OCR fallback and Gemini LLM to detect missing clauses and rank keywords. Handles heavy compute tasks asynchronously.
Python
OCR
LLM Agents
Celery
Redis
MongoDB
Next.js

Multimodal LLM framework for generating Arduino circuits. Fine-tuned Qwen models with LoRA to achieve 75% accuracy on complex hardware generation tasks.
PyTorch
Hugging Face
LoRA/QLoRA
RAG
Python

A comprehensive educational platform serving 5,000+ monthly active users. Handles 150,000+ monthly API requests with sub-100ms response times.
React Native
Node.js
Redux
MongoDB
Firebase
AWS
ShadCN