Applied AI Student Researcher

ServiceNow

• Working on benchmarking content moderation in enterprise-level agentic systems.
• Focus on long-context multi-modal agentic systems with reasoning-oriented content moderation models.

Research Intern

Infosys

• Worked on the red teaming of Large Language Models (LLMs) via membership inference and data extraction attacks.
• Investigated the privacy implications associated with model merging techniques.

Undergraduate Researcher

Center for Security, Theory & Algorithmic Research (CSTAR), IIIT Hyderabad

• Investigated ownership verification and watermarking for Deep Neural Networks (DNNs).
• Worked on watermarking Graph Neural Networks (GNNs) for the Link Prediction (LP) task.
• Utilized PyTorch Geometric to evaluate our proposed watermarking scheme across 4 model architectures and 8 real-world datasets.

Research Assistant

CSTAR, IIIT Hyderabad

• Worked on various ML security problems including Explainable AI (XAI), Explanation-aware Backdoor attacks, and Adversarial attacks.