Curriculum Vitae

Invited Reviewer, SocialAgent Workshop @ NeurIPS 2026 • 1 of 23 global case studies selected from 235+ submissions across 50+ countries, India AI Impact Summit 2026 (MeitY × UN Women) • Author, Springer LNNS (ICICC 2026), top 15% of papers reviewed • 36,000+ cumulative downloads across open-source AI infrastructure (PyPI) • Advisor, Machine Learning Tech Society (Dean-invited)

AI/ML Delivery Intern, Virtusa Consulting Services Pvt Ltd

Selected through Virtusa's Jatayu Hackathon 2026, Agentic AI track. Built CodePerfect Auditor end-to-end: a clinical NLP extraction layer, configurable ICD-10/CPT code-mapping logic, and a discrepancy-detection engine with revenue-impact estimation, reconciling clinical documentation against billed codes. Rendered every flagged discrepancy through an interactive Clinical Evidence Graph tracing it back to supporting evidence. Shipped as a production-style FastAPI service with a Streamlit reviewer interface.

Undergraduate Thesis — Explainable Multimodal AI

Saveetha Engineering College, India.
Designed an explainable vision–language generative framework for radiology report synthesis, coupling generative VLMs with structured clinical grounding to produce interpretable, evidence-linked reports. Owned the model architecture, evaluation protocol, and interpretability analysis end-to-end. Manuscript in preparation for journal submission.

B.E. in Electronics and Communication Engineering

Saveetha Engineering College, India. CGPA: 8.10/10.
Coursework: Artificial Intelligence, Deep Learning, Natural Language Processing, Random Processes and Statistics, Prompt Engineering.

Languages: Python • ML/DL: PyTorch, Vision Transformers, Transformer-based LLMs/SLMs • AI Systems: Retrieval-Augmented Generation, Explainable AI (XAI), Clinical & Legal NLP • Deployment: FastAPI, Streamlit • Tooling: Hugging Face, PyPI, Git/GitHub

Multimodal AI • Vision–Language Models • Explainable AI • Constitutional AI • AI Governance • Public Interest Technology • Healthcare Intelligence • Retrieval-Augmented Generation • Digital Policy

NariRaksha-3B — Gender-Responsive Safety Model

Released a 3B-parameter safety model built on the NariRaksha research, packaged for public deployment and downstream fine-tuning.

Explainable Vision–Language Model for Radiology

Integrated Vision Transformers and BioGPT into a multimodal clinical intelligence system for radiological reasoning, structured report generation, and confidence-aware interpretability.

Open-Source Constitutional Intelligence Platform

Built a semantic retrieval and structured-analysis platform for India's constitutional corpus. Surpassed 36,000 cumulative downloads on PyPI.

Indic Constitutional & Legal Reasoning Benchmark

Constructed a benchmark dataset for evaluating constitutional and legal reasoning in Indic-language NLP systems.

Explainable SLM-Guided Vision–Language Model for Multi-Class Skin Lesion Recognition

Springer LNNS, Proceedings of ICICC 2026. Indexed in Scopus and Web of Science. Recognized among the top 15% of papers reviewed.

NariRaksha: Gender-Responsive AI for Women's Safety

India AI Impact Summit 2026 Compendium, UN Women & MeitY, Government of India. Selected among 23 global case studies from 235+ submissions across 50+ countries.

IndianConstitution: Trustworthy, Reproducible, and Type-Safe Infrastructure for Legal NLP

Zenodo Preprint.

Beyond Code Smell: Measuring Architectural Conformance Drift in Long-Horizon Autonomous Coding Agent Sessions — A Pilot Study

Zenodo Preprint.

Invited Reviewer, SocialAgent Workshop @ NeurIPS 2026

Second Workshop on Large Language Models for Social Reasoning and Simulation. Peer reviewing submissions via OpenReview for the workshop's review cycle, September 7–23, 2026.

Advisor, Machine Learning Tech Society

Invited by the Dean of Information and Communication Technology to lead the institution's machine learning community. Designed and delivered advanced programs on AI Agents, Quantum Machine Learning, and Financial Machine Learning while mentoring student researchers.

Published Contributor, India AI Impact Summit 2026

Lead author selected among 23 case studies from 235+ submissions across 50+ countries. Individually recognized by Sudeshna Mukherjee, Head of Communications, UN Women India Country Office, at the official publication launch.

First Prize, Department of ECE Project Contest

Saveetha Engineering College. Awarded for the radiology VLM thesis work; certified by the Head of Department, ECE.