No image available...
My Portfolio Website
2023
• Personal project
Description
Built with Next.js, TypeScript, Tailwind CSS, shadcn/ui, and Framer Motion, deployed on Vercel. Features a Hybrid RAG System leveraging all-mpnet-base-v2 for semantic search, onnxruntime-web for optimized inference, and llama-3.3-70b-versatile for AI-driven text generation. Additionally, it integrates the GitHub API for real-time project updates, ensuring dynamic content. The system enhances user interactions with intelligent query processing, intent classification, and adaptive responses powered by fine-tuned embeddings and efficient retrieval methods. It also has a terminal feature, allowing users to run commands, enhancing interactivity and user engagement.
Platform
Web
Key features
- AI-powered semantic search
- Hybrid RAG system with intelligent retrieval
- Interactive terminal emulator with basic command execution (
ls,clear,exit,whoami, etc.) - Built-in PDF viewer for resume preview
- Optimized AI inference using ONNX Runtime and Groq
- Text generation capabilities with LLaMA-3.3-70B-Versatile
- Fine-tuned intent classification for better query understanding
Technologies used
NextJS
React
Typescript
Tailwind
Framer motion
shadcn/ui
Vercel
Github API
Python
Python Virtual Environment
all-mpnet-base-v2
llama-3.3-70b-versatile
Groq
ONNX Runtime
Scikit-learn
PyTorch