cv

MSc. (Hons.) in Biological Sciences & BEng. (Hons.) in Electronics & Instrumentation

BITS Pilani, K.K. Birla Goa Campus

Minor in Data Science in Climate & Health

• Linear Algebra
• Probability & Statistics
• Applied Statistical Methods
• Foundations of Data Science
• Machine Learning
• Object Oriented Programming
• Bioinformatics
• Operating Systems

AI Research Intern

CSIR - Central Electronics Engineering Research Institute (CEERI)

Worked on brain-computer interfaces using EEG signals and deep learning.

• Conducted exploratory analysis on multi-channel EEG time-series data to extract temporal correlations and developed interactive visualization dashboard
• Built lightweight CNN classifier with attention mechanisms for real-time EEG finger movement detection, achieving 85% accuracy and F1-score of 0.75

Undergraduate ML Researcher

APPCAIR - BITS Pilani

Researching applications of machine learning and LLMs for drug discovery and computational biology.

• Developed semi-automatic system for molecule generation and retrosynthesis leveraging LLMs (GPT-4, Claude) with logical feedback loops, generating 20+ novel protein inhibitor candidates targeting JAK2 and dopamine beta-hydroxylase (DBH)
• Designed and implemented LSRPI (Location-Specific RNA-Protein Interaction) Transformer-based model to predict residue-level RNA-protein interactions and generate interaction matrices, validated through docking simulations with Gnina
• Improved XGBoost-based RNA-protein interaction model, achieving 5% accuracy boost and 10% reduction in processing time through optimized preprocessing

ML Research Intern

Deep Forest Sciences

Working on machine learning applications for drug discovery and materials science.

• Architected end-to-end image preprocessing pipeline for Prithvi drug discovery platform and deployed to production for client use, implementing automated object counting and UNet-based segmentation
• Researched generation of novel low-k dielectric materials using LLMs and using Molecular Dynamics and DFT simulations for validation, screening 40+ candidate materials and reducing screening time by 83% (30+ days to 5 days) with 15-20% success rate

Open Source Contributor

DeepChem

Contributing to DeepChem, an open-source Python library for deep learning in drug discovery, materials science, quantum chemistry, and biology.

• Added 7+ models including SCScore, UNet, OneFormer, and Protein Language Models (ESM-2, ProtBERT) to DeepChem with PyTorch, expanding capabilities in synthetic feasibility, segmentation, and protein analysis for 40k+ users
• Developed GPU-accelerated parallel ODE solver infrastructure using PyTorch, enabling highly scalable neural ODE training and inference
• Optimized image loading pipeline and authored educational tutorials on automated microscopy and cell segmentation, resulting in 10% boost in site traffic

Visiting Research Fellow

Brown University - Serre Lab

Developing embodied AI agents for visual perspective-taking and mental simulation tasks, learning to predict field-of-view from different spatial positions through active interaction.

• Designing multimodal ML systems that fuse vision, language, and action signals to enable cross-task generalization
• Creating custom simulation datasets in Isaac Lab and applying RL using OpenAI Gym and PyTorch to build scalable prototypes for robotics applications

6-DOF Robotic Arm with RL-Based Digital Twin

Designed and built 6-DOF robotic arm using servos, stepper motors, and ESP32 microcontroller with gripper, wrist (pitch/yaw), elbow, shoulder, and rotating base joints. Developed digital twin simulation for RL-based training on object localization and grasping using image and sensor data for movement control and task execution.

• Python
• PyTorch
• ESP32
• OpenCV
• Reinforcement Learning
• Robotics

Time-Masked Autoencoders for Fluid Dynamics

Collaborated with Imperial College London researchers to study temporal masking in video autoencoders for fluid dynamics simulations. Implemented models predicting up to 20 future frames of Shallow Water simulations with up to 80% input masking, maintaining 80%+ SSIM.

• Python
• TensorFlow
• PyTorch
• Video Autoencoders

Mixture of Experts for Named Entity Recognition

Implemented MoE layer from scratch and integrated into BiLSTM model for CoNLL 2003 NER dataset, achieving 12% accuracy increase and 32% F1-score improvement over baseline.

• Python
• PyTorch
• HuggingFace
• Named Entity Recognition

Autonomous Rock Detection System for Mars Rover

Led development of computer vision system for autonomous life detection, including transfer learning-based multi-class rock classifier for biological significance assessment. Achieved 2nd place at Anatolian Rover Challenge (Turkey) and won Excellence Award at International Rover Challenge (India).

• Python
• TensorFlow
• Linux
• Computer Vision
• Transfer Learning

Compact and Efficient RNA Representations with RNAvec Enable Residue-Level Interaction Mapping

In preparation for Nucleic Acids Research (NAR)

Novel representation learning approach for accurate residue-level RNA-protein interaction prediction. Authors: Omkar S. Sathe, Aaron R. Menezes, Sanket R. Gupte, Aman A. Kattuparambil, Ashwin Srinivasan, Raviprasad Aduri

Open Source Infrastructure for Automatic Cell Segmentation

Arxiv Preprint

Authors: Aaron R. Menezes, Bharath Ramsundar

Open Source Differentiable ODE Solving Infrastructure

AAAI Workshop on AI to Accelerate Science and Engineering (AI2ASE)

Authors: Rakshit Singh, Aaron R. Menezes, Rida Irfan, Bharath Ramsundar

INSPIRE Scholar

Department of Science and Technology, Government of India

Awarded to top 1% of science students in Class 12 HSC board exams with scholarship grant to support further studies in science

Google Summer of Code (GSoC) Mentor

Google

Mentoring at DeepChem for Target Conditioned Antibody Sequence Generation Using Protein Language Models

2nd Place at Anatolian Rover Challenge (ARC)

Anatolian Rover Challenge, Turkey

Led multidisciplinary team of 7 in designing rover for autonomous life detection in extreme terrain

Excellence Award at International Rover Challenge

International Rover Challenge, India

Won Excellence Award for autonomous rock detection system