Hi. I'm Rishab Bothra.

I am a Data Science student at the University of Marburg, currently bridging the gap between academic theory and industrial application as an Intern/Thesis student at Belden. As an aspiring Generative AI Engineer, I am passionate about building autonomous systems and streamlining workflows using LLMs and AI Agents. I specialize in leveraging the AWS ecosystem to create scalable, real-time solutions that solve complex automation challenges. I’m driven by the potential of GenAI to transform how we work and am always looking for new ways to push the boundaries of what's possible with intelligent agents.

Learn more about me.

My Working Experience

Gen AI Intern

Designed and architected an end-to-end Generative AI-powered system to automate Git patch conflict resolution, significantly reducing manual developer intervention. By implementing a RAG-based architecture, I enabled the model to leverage historical context from git blame and reviewer feedback to make highly accurate, context-aware code corrections. I further optimized the system by introducing function-level refactoring for complex code blocks and building a comprehensive tracking module to monitor success rates, latency, and LLM performance. This resulted in a self-evolving workflow that captures developer feedback into a structured knowledge base for continuous learning.

Here's my Research work

Diabetes Prediction using Machine Learning Algorithms

In this paper we have proposed a diabetes prediction model using Machine Learning algorithm for better classification prediction. We have tried different Machine Learning algorithms to find which gives the better accuracy of classification.
Paper
Code

Deep Learning Models for Electricity Consumption Prediction

This paper analyzes and compares deep learning models, including LSTM and GRU, for electricity consumption forecasting, demonstrating improved accuracy through hybrid approaches. It focuses on applications in sustainable energy management and is under review with Elsevier.

Here’s some project I made recently.

Live Face Mask Detection

Considering the COVID situation I have tried to build an amazing Computer vision project using DNN ( self build CNN Architecture). In this project we can detect LIVE whether the person is wearing mask or not.

Digit classification using SVD

In this project we have tried to recognize handwritten digit (0 - 9). The dataset was imported from sklearn dataset library i.e mnist dataset. We built end to end LogisticRegression model to predict handwritten digit .

Text Generator using GPT2

The project is simple we give input as a phrase or sentence or question and as an output we get a small essay generated by this application. Since by name we can understand it is a text generator application.

Ongoing Project

Fundus Disease Recognition

Collaborating in a team of 3. Project initiated in Summer Semester 2024. Supervised by Daniel Schneider, Ph.D. (under Prof. Dr. Bernd Freisleben, Dean of the University).

Measurement of the Thickness of the Uppermost Skin

Collaborative project with AI Lab Marburg. Working in a team of 3 with a mentor. Scheduled to start in December 2024 and estimated to complete by March 2025.