Projects

AdaptAI: A Personalized Solution to Sense Your Stress, Fix Your Mess, and Boost Productivity

AdaptAI is a multimodal AI system that enhances productivity and well-being by tailoring interventions to individual needs. Unlike generic tools, AdaptAI integrates egocentric vision, audio, physiological signals, and LLM-driven workflows to provide context-aware support. Its Tone-Adaptive Conversational Agent enhances productivity by adjusting responses to be more positive and supportive during stressful moments.

Energy Efficient Navigation for ASVs in Turbulent Vortical Fields

Developed RL Models to optimize the ASV’s trajectory and control mechanisms to minimize energy expenditure, relying predominantly on thrust-based actuation while simultaneously exploiting naturally occurring environmental phenomena, such as turbulence and vortex structures.

KhalasiIO: Detecting Situational Impairments With Reasoning-based Memory Bank

Detects and addresses temporary impairments caused by environmental factors like noise, lighting, and stress. Integrates wearable devices and contextual memory for personalized, real-time interventions, reducing cognitive load and frustration. Promising results highlight scalability and improved accessibility.

PosePilot: A Novel Posture Correction System Leveraging BiLSTM and Multihead Attention

PosePilot: provides corrective feedback on yoga poses by using LSTM networks to detect poses and offer posture guidance. It analyzes body key points from videos to predict poses and forecast corrections, ensuring proper technique and posture for practitioners. The system helps overcome challenges like time, cost, and lack of instruction, offering personalized, real-time feedback for improved practic

EleSpec

Objective of the software EleSpec is to simplify the process of visualization and exploration of synthetic spectra in an intuitive user interface, facilitating the detection and identification of atoms and molecules. EleSpec offers users the liberty to input spectra in any text extension of their choice. Whether it be CSV, TXT, or other common formats, the software seamlessly integrates the data. Furthermore, EleSpec supports xz compression, allowing users to efficiently manage and analyze large datasets without sacrificing performance or storage.

AstroMCMC

I worked on this project during my internship at Physical Research Laboratory (PRL) with Dr. Arvind Singh Rajpurohit. The project involved analyzing data from the VLT (Very Large Telescope) telescope at ESO (European Southern Observatory). I used a complex statistical method called Monte Carlo Markov Chain (MCMC) to initially fine tune a model parameter and later to know which metal elements were abundantly present on the Red Dwarf Star.

Therepia: Parkinson Disease Tracking App

Therepia mobile app helps monitor Parkinson's disease by using advanced sensors like accelerometers and gyroscopes to track vibrations and movement. With Machine Learning, it analyzes patterns to understand how medication affects symptoms, offering an innovative, home-based solution.

page2circuit

page2circuit converts a hand-drawn circuit into digital schematics. It was made using PyTorch and OpenCV. The model was trained using open dataset consisting about 14,000 hand-drawn images of 11 distinct components. The model achieved validation accuracy of 90.35% and training accuracy of 99.9 %. To make it accessible we decided to host using Streamlit, which extremely sharing friendly.

Ekaant

An application that guides you through your meditation practise. Ekaant offers individuals a platform to monitor their regular meditation progress. With the option to play soothing music in the background, it aids in developing a comfortable frame of mind for meditation. The app was created using the Flutter framework, which is accessible to Android and iOS users.

Rushiraj Gadhvi