Harshit V

Harshit V

I'm a Tech enthusiast skilled in Generative AI, Machine Learning, and WordPress development. I love developing projects and becoming a mid-polymath background character is the end-goal of my life I guess.


Previously Interned at Game Plan Tribe as WordPress Freelancer, Prag Robotics as Machine Learning Engineer. I did my B.Tech in AI/ML at SRM University, Chennai.


Portfolio

I'm interested in training deep learning models, solving real-world problems with generative AI, and developing useful Full Stack web apps.

LawGPT

LawGPT: RAG based Generative AI Attorney Chatbot

RAG
LangChain
Streamlit
Python

LawGPT is an AI attorney chatbot built with Streamlit, LangChain, and TogetherAI, trained on Indian Penal Code (IPC) data. Ask any legal questions, and it will provide accurate justice based on the IPC.

RAGLess

RAG without Vector-DB? - Yes

RAG
LangChain
Streamlit
Python

This project breaks away from traditional methods using embeddings and vector databases by employing a TF-IDF retriever, long context reordering, and a flash reranker. This approach improves retrieval speed with k=8 while maintaining accuracy comparable to vector databases.

MedChat

MedChat - RAG based AI Chatbot for 🇮🇳 Medicines

RAG
LangChain
Streamlit
Python

MedChat is a RAG-based AI chatbot for inquiries about Indian medicines. Built with Streamlit, LangChain, and Mistral LLM, it uses open web data on Indian medicines to provide accurate responses to user questions about medications.

Time Series LSTM

Time Series Forecasting LSTM Streamlit

LSTM
Pytorch
Streamlit
Python

Time Series Forecasting based on Streamlit GUI. Featuring pre-built data processing, correlation visualization, and customizable models, it supports both univariate and multivariate forecasts. Supports customizable LSTM features and interactive Plotly graphs.

AI-Stock Analyzer

AI-Stock Analyzer

LangChain
ScreenerAPI
Streamlit
Python

AI-Stock-Analyzer is a generative AI webapp for performing fundamental analysis of the NSE stock market data. This project was developed using Streamlit LangChain and Gemini Flash LLM. Used selenium python to webscrape screencer website for stock data and to create API.

Go to Top ⬆️