about me

Machine Learning Engineer 1

FactSet Research Systems

• Currently working on an anomaly-detection framework to flag outlier trades and high-risk brokers, separating macro-driven anomalies from broker-specific behavior using statistical tests.
• Built and shipped to production an end-to-end RAG application that lets users register securely using a cert and automatically deploy personalized RAG systems on their own data.
• Implemented Redis-based job scheduling, ElasticSearch VectorDB with re-ranking for improved retrieval accuracy, and a multi-tenant architecture enabling non-technical users to create production-ready RAG apps effortlessly.

Research Assistant

Mohamed bin Zayed University of Artificial Intelligence

• Currently working with Dr. Jian kang to develop privacy and utility preserving synthetic temporal relational databases.

Research Assistant

Center for Security, Theory and Algorithmic Research

• Worked with Dr. Ankit Gangwal to develop a partial model extraction attack to steal a DNN classifier and effectively evade watermark detection without losing utility.
• Completed a paper that got accepted into AsiaCCS 2025, presented as a poster in Hanoi, Vietnam.

Software Engineer Intern

Center for Visual Information Technology, IIIT Hyderabad

• Collaborated with Dr. Ravi Kiran Sarvadevabhatla to develop a robust manuscript annotator web application using Mirador, React, Cantaloupe image servers, and IIIF image API.
• Additional features include dynamic image annotations, and text transliteration for annotating in Indic languages.

Honors Student and Undergraduate Research Assistant

Machine Learning Lab, IIIT Hyderabad

• Collaborated with Dr. Charu Sharma and Dr. Ankit Gangwal to develop two frameworks: GENIE, a robust and secure approach for watermarking Graph Neural Networks for link prediction tasks, and LoReTTA, a novel adversarial poisoning technique designed for Temporal Graph Neural Networks on Continuous-Time Dynamic Graphs.

Bachelor of Technology

IIIT Hyderabad

Computer Science

Research Excellence Award

IIIT Hyderabad

For exceptional research contributions and high-impact publications in top-tier conferences.

KVPY Fellowship

Department of Science and Technology, Government of India

Awarded the KVPY Fellowship with All India Rank (AIR) 193 for demonstrated aptitude in scientific research.

Joint Entrance Examination (JEE)

Achieved All India Rank (AIR) 893 in JEE Main and AIR 1,933 in JEE Advanced.

LoReTTA: Low Resource Framework To Poison Continuous Time Dynamic Graphs

AAAI Conference on Artificial Intelligence

Temporal Graph Neural Networks (TGNNs) are increasingly used in high-stakes domains, such as financial forecasting, recommendation systems, and fraud detection. However, their susceptibility to poisoning attacks poses a critical security risk. We introduce LoReTTA (Low Resource Two-phase Temporal Attack), a novel adversarial framework on Continuous-Time Dynamic Graphs, which degrades TGNN performance by an average of 29.47% across 4 widely benchmark datasets and 4 State-of-the-Art (SotA) models. LoReTTA operates through a two-stage approach: (1) sparsify the graph by removing high-impact edges using any of the 16 tested temporal importance metrics, (2) strategically replace removed edges with adversarial negatives via LoReTTA’s novel degree-preserving negative sampling algorithm. Our plug-and-play design eliminates the need for expensive surrogate models while adhering to realistic unnoticeability constraints. LoReTTA degrades performance by upto 42.0% on MOOC, 31.5% on Wikipedia, 28.8% on UCI, and 15.6% on Enron. LoReTTA outperforms 11 attack baselines, remains undetectable to 4 leading anomaly detection systems, and is robust to 4 SotA adversarial defense training methods, establishing its effectiveness, unnoticeability, and robustness.

GENIE: Watermarking Graph Neural Networks for Link Prediction.

Transactions on Machine Learning Research

The rapid adoption, usefulness, and resource-intensive training of Graph Neural Network (GNN) models have made them an invaluable intellectual property in graph-based machine learning. However, their wide-spread adoption also makes them susceptible to stealing, necessitating robust Ownership Demonstration (OD) techniques. Watermarking is a promising OD framework for deep neural networks, but existing methods fail to generalize to GNNs due to the non-Euclidean nature of graph data. Existing works on GNN watermarking primarily focus on node and graph classification, overlooking Link Prediction (LP). In this paper, we propose Genie (watermarking Graph nEural Networks for lInk prEdiction), the first scheme to watermark GNNs for LP. Genie creates a novel backdoor for both node-representation and subgraph-based LP methods, utilizing a unique trigger set and a secret watermark vector. Our OD scheme is equipped with Dynamic Watermark Thresholding (DWT), ensuring high verification probability while addressing practical issues in existing OD schemes. We extensively evaluate Genie across 4 diverse model architectures (i.e., SEAL, GCN, GraphSAGE and NeoGNN), 7 real-world datasets and 21 watermark removal techniques and demonstrate its robustness to watermark removal and ownership piracy attacks. Finally, we discuss adaptive attacks against Genie and a defense strategy to counter it

Disappearing Ink: How Partial Model Extraction Erases Watermarks

AsiaCCS

Deep neural networks have become invaluable intellectual property in machine learning. To deter model theft, Watermarking has emerged as a prominent defense by embedding hidden “trigger sets” that aid in ownership verification. However, current watermarking solutions primarily address scenarios where adversaries steal the entire model. In this paper, we reveal a critical gap: partial model extraction, where only a subset of classes is stolen, substantially degrading the watermark’s reliability. We introduce two attacks, Partial Model Extraction and Partial Knowledge Distillation, which reduce watermark accuracy by up to 80% while retaining strong performance on the stolen classes. Through extensive experiments on CIFAR10 and CIFAR100 against two state-of-the-art watermarking schemes, we demonstrate the need for more robust watermarking strategies that resist partial-class theft.

PatchPoison: Poisoning Multi-View Datasets to Degrade 3D Reconstruction

SPAR-3D Workshop at CVPR

3D Gaussian Splatting (3DGS) has recently enabled highly photorealistic 3D reconstruction from casually captured multi-view images. However, this accessibility raises a privacy concern: publicly available images or videos can be exploited to reconstruct detailed 3D models of scenes or objects without the owner’s consent. We present PatchPoison, a lightweight dataset-poisoning method that prevents unauthorized 3D reconstruction. Unlike global perturbations, PatchPoison injects a small high-frequency adversarial patch, a structured checkerboard, into the periphery of each image in a multi-view dataset. The patch is designed to corrupt the feature-matching stage of Structure-from-Motion (SfM) pipelines such as COLMAP by introducing spurious correspondences that systematically misalign estimated camera poses. Consequently, downstream 3DGS optimization diverges from the correct scene geometry. On the NeRF-Synthetic benchmark, inserting a 12 X 12 pixel patch increases reconstruction error by 6.8x in LPIPS, while the poisoned images remain unobtrusive to human viewers. PatchPoison requires no pipeline modifications, offering a practical, drop-in preprocessing step for content creators to protect their multi-view data.