Education
The George Washington University
Master of Science in Computer Science
August 2024 - May 2026
Washington, D.C.
PES University
Bachelor of Technology in Computer Science and Engineering
December 2020 - May 2024
Banglore, India
Experience
AI Fellow
Handshake
•Designed and refined prompts for large language models (LLMs) evaluation, focusing on reasoning quality, bias detection, and instruction adherence.
•Performed human-in-the-loop evaluation of model outputs, performance and behavior for accuracy, fairness, consistency, and domain-specific reasoning.
•Annotated large-scale text and video data, including detailed frame-level and cross-video analysis to capture visual changes and motion patterns across sequences.
December 2025 - Present
United States (Remote)
Research Assistant
PES University
•Developed a CNN-Transformer multimodal model to generate natural language metaphors from images by learning source-target domain mappings, and utilizing cross-modal attention to align visual features with semantic concepts.
•Applied multi-task learning with metaphor type, sentiment, and intent to improve generation quality, evaluated model performance using BLEU and BERTScore, along with qualitative analysis of interpretability, and semantic alignment.
June 2023 - December 2023
Banglore, India
Skills
Programming & Scripting
- Python
- R
- SQL
- Java
- JavaScript
Machine Learning & AI
- scikit-learn
- TensorFlow
- PyTorch
- XGBoost
- MLflow
- Bedrock
- SageMaker
Data Analytics & Visualization
- NumPy
- pandas
- Matplotlib
- Seaborn
- Tableau
- Power BI
Data Engineering & Cloud
- Spark
- Airflow
- Kafka
- Redshift
- EC2
- S3
- Aurora
- Lambda
- MySQL
- PostgreSQL
- MongoDB
- Neo4j
Tools & DevOps
- Git
- Docker
- Kubernetes
- Jenkins
- Jira
Projects
BiLSTM-CNN With Attention For Remaining Useful Life Prediction In Turbofan Engines
Hybrid deep learning model combining BiLSTM, Transformer, and CNN architectures to estimate the Remaining Useful Life (RUL) of aircraft engines using the NASA C-MAPSS dataset. The model incorporates CBAM attention to enhance focus on key sensor features, positional encoding to preserve the temporal structure of sensor sequences, residual dilated convolutions to capture long-range dependencies, and transformer blocks for contextual sequence modeling, achieving average RMSE of 18.88.
- TensorFlow
- scikit-learn
Crowd Behavior Anomaly Detection in Surveillance Videos
A weakly supervised Multiple Instance Learning (MIL) framework for crowd behavior anomaly detection in surveillance videos, leveraging I3D-based spatio-temporal features extracted from RGB and optical flow streams. Utilizes multi-stream neural architecture, incorporating attention, residual connections, dropout, and layer normalization, achieving an AUC of 0.9022 and outperforming baseline classifiers in precision-recall balance.
- PyTorch
Spatiotemporal Analytics and Pattern Mining in Urban Complaints
Data mining and analytics on NYC 311 service request data, covering large-scale data cleaning, contrast and sequential pattern mining, anomaly detection for complaint spikes, and predictive modeling for service resolution time. The project applies statistical and machine learning models to identify patterns in complaint volume, types, and spatial distribution.
- scikit-learn
- NumPy
- pandas
AI-Powered Diagnostic Support Tool
Retrieval-Augmented Generation (RAG) system over PubMed records and the OpenAI API to generate disease predictions, medical test and treatment recommendations from patient symptoms, demographics, and medical history, incorporating agentic AI for context-aware reasoning and improved diagnostic accuracy.
- Node.js
- React
Vehicle Detection and Classification
SSD-Mobilenet V1-based traffic monitoring system to detect, count, and classify vehicles in real-time, addressing need for efficient traffic control. Implemented vehicle color and speed prediction, storing images of detected vehicles, and logging detection data to a CSV file for analysis.
- TensorFlow
- OpenCV
Evaluating Traditional and Ensemble Learning Techniques for Churn Prediction
Built and evaluated multiple machine learning models to predict customer churn using a telecom dataset. Applied feature engineering and preprocessing, and benchmarked classifiers including Logistic Regression, SVM, Decision Trees, Random Forest, Gradient Boosting, and ensemble methods.
- pandas
- NumPy
- scikit-learn
Certifications
AWS Certified AI Practitioner
Issued: September 2025 | Expires: September 2028
AWS Certified Cloud Practitioner
Issued: September 2025 | Expires: September 2028