Prateek Chavala

Prateek Chavala


Skills
Java
machine learning
transformers
MongoDB
sq\
Projects

Scene Solver

Crime scene analysis app that detects and classifies harmful objects in images & video using Vision Transformer + YOLOv8

Frontend: React.js
Axios
React Router Backend: FastAPI (Python) Database: MongoDB Models: Vision Transformer
YOLOv8
distilgpt2 Tools: Git

LegalSeg

Unlocking the Structure of Indian Legal Judgements Through Rhetorical Role classification. The primary problem addressed in this work is the automatic segmentation of Indian legal judgments into rhetorical roles such as Facts, Issue, Arguments, Reasoning, Decision, and None. Legal judgments are lengthy, complex, and unstructured, making it difficult for lawyers, researchers, and citizens to extract key information efficiently. Traditional methods and small-scale datasets have proven insufficient for capturing the intricacies of legal language and the sequential flow of arguments. As a result, there is a strong need for a large annotated dataset and advanced NLP models that can accurately classify sentences into rhetorical roles. Solving this problem will improve legal research, enable better summarization, and support the development of automated decision-making systems in the legal domain

Frontend React.js Axios React Router Framer Motion Backend Node.js Express.js MongoDB Mongoose JWT Authentication ML / NLP Python-based processing Transformer-based classification
Education

Btech

Keshav Memorial College of Engineering

Sept 2023 to May 2027

Jr College

Narayana

Jun 2021 to May 2023