Back to project
Portfolio
CompletedLiveGithub

Portfolio

My personal portfolio — built to showcase projects, blogs, and experiments with intentional design and clean architecture.

Overview

V2 Portfolio is the second iteration of my personal website, built from scratch with a focus on performance, modularity, and clean UI.

It serves as a central hub for everything I'm working on — projects, blog posts, my coding setup, movies I enjoy, and random things I find interesting.

No templates. No fluff. Just purposeful design backed by solid engineering.

Tech Stack

  • Next.js 14 (App Router)
  • React 18
  • TypeScript
  • Tailwind CSS
  • shadcn/ui
  • Magic UI

Why I Built This

My first portfolio felt limiting — hard to extend, hard to maintain, and not really me.

I wanted to rebuild it with a proper architecture: feature-based folder structure, a blog system that actually works, and a design that reflects how I think about code and interfaces.

This project pushed me to get serious about Next.js App Router, MDX-based content systems, and building UI components that feel polished without being bloated.

AI & RAG Experience

Beyond frontend development, I have also worked with Retrieval-Augmented Generation (RAG) systems and modern AI application architecture.

For personal experiments and AI-powered projects, I implemented complete RAG pipelines using embedding models and vector databases to enable intelligent document retrieval and context-aware responses.

What I Worked With

  • Retrieval-Augmented Generation (RAG)
  • Text Embeddings
  • Vector Databases
  • Semantic Search
  • Context Retrieval Pipelines
  • Document Indexing & Chunking
  • AI-powered Knowledge Base Systems

Key Learnings

Building RAG systems helped me understand how modern AI products retrieve relevant information instead of relying solely on model memory. I explored the complete workflow:

  1. Document ingestion and preprocessing
  2. Text chunking strategies
  3. Embedding generation
  4. Vector storage and indexing
  5. Similarity search and retrieval
  6. Context injection into LLMs
  7. Response generation with retrieved knowledge

This gave me practical experience in designing AI systems that are scalable, accurate, and capable of working with custom datasets.

Why It Matters

Working with RAG and vector databases expanded my skillset beyond traditional web development and introduced me to real-world AI engineering concepts used in modern products such as AI assistants, knowledge bases, document search systems, and customer support automation.

Features

  • Dynamic Hero Section
  • Animated UI components
  • Projects showcase with MDX content
  • Blog system
  • Movie section
  • Personal coding setup page
  • Social links integration
  • Feature-based modular architecture
  • Fully responsive design

Project Structure

src/
├── app/          # App Router pages
├── components/   # Reusable components
├── lib/          # Utility functions
├── data/         # Static/minor data
└── features/     # Feature-based UI sections

Current Status

  • Live at tanishtirpathi.me
  • Actively maintained with continuous UI and content updates
  • 117+ commits and growing

Future Plans

  • Expanded blog with more technical writing
  • More project case studies
  • Performance optimizations
  • Additional interactive sections
  • Ongoing design refinements
Back to project