Production GenAI and RAG
RAG assistants, multilingual patient support, knowledge retrieval, evaluation workflows, and lead-aware healthcare communication systems.
I build production AI systems for healthcare and study how to make them clinically responsible.
AI Decision Science Analyst at Accenture with experience across diagnostic AI, GenAI and RAG systems, OCR automation, medical imaging research, and healthcare AI governance.

I am an AI Decision Science Analyst at Accenture, working across decision science, GenAI, analytics, and AI-led transformation use cases. My practical focus is healthcare and life sciences AI, especially systems that move from prototype to real operational workflows.
Previously at Dr. Lal PathLabs, I built patient-facing GenAI systems, RAG workflows, OCR and document intelligence automation, compliance verification, demand forecasting, MIS automation, and diagnostic interpretation pipelines.
At VEDAs Lab, I contribute to multi-institution medical AI research across microscopy, ultrasound, retinal imaging, neuroimaging, robust machine learning, and reproducible evaluation workflows.
I founded the Foundation for Ethical AI in Healthcare to support evidence-based, clinically responsible AI adoption through research, policy dialogue, standards, and community engagement.
A practical mix of healthcare AI engineering, medical AI research, and responsible deployment.
RAG assistants, multilingual patient support, knowledge retrieval, evaluation workflows, and lead-aware healthcare communication systems.
Research and implementation across microscopy, ultrasound, retinal imaging, neuroimaging, segmentation, multimodal learning, and model evaluation.
OCR preprocessing, Azure Document Intelligence, structured extraction, compliance verification, and workflow automation for diagnostic operations.
Responsible AI adoption, standards, policy dialogue, clinical validation, safety, accountability, and evidence-based deployment in healthcare.
Working with AI and Data, Strategy and Consulting teams on decision science, GenAI, analytics, and AI-led transformation use cases for enterprise clients.
Built production healthcare AI systems across GenAI support, RAG assistants, OCR automation, compliance verification APIs, MIS reporting, forecasting, and diagnostic interpretation workflows.
Contributing to multi-institution medical AI research across microscopy, ultrasound, retinal imaging, neuroimaging, robust ML, retrieval systems, and reproducible evaluation pipelines.
Contributed to BrainLM, a foundation model trained on 6,700 hours of fMRI recordings for cognition and behaviour decoding from brain activity.
Production and research systems from the resume, structured so more projects can be added from the content file.
Multilingual GenAI and RAG support across web and patient app channels, handling 3,000+ patient interactions per day and generating about 400 leads per day.
Learn moreDoctor-facing RAG assistant with lead capture and centralized sales app tracking for improved query coverage and end-to-end visibility.
Learn moreAzure Document Intelligence and preprocessing pipeline processing 500+ forms per month and reducing manual effort by 20+ hours per month.
Learn morePatient-level diagnostic interpretation framework using reference intervals, Mahalanobis distance, percentile scoring, lifestyle scoring, consistency checks, and adaptive weighting.
Learn moreA Section 8 healthcare AI initiative focused on ethical, evidence-based, and clinically responsible AI adoption.
Learn moreEach entry supports a title, authors, venue, year, summary, link, and image path for future paper cards.
Research contribution in robust machine learning and high-dimensional data analysis. Add the final title, DOI, PDF, and project image when public.
First/equal-first author work on medical microscopy and inverse protocol prediction for spheroid experiments.
Foundation model work using 6,700 hours of fMRI recordings to support brain activity representation learning and cognition decoding.
A blog-ready section for healthcare AI essays, project notes, and publication explainers.
A practical note on clinical validation, monitoring, accountability, and why benchmark accuracy is not enough for patient-facing AI.
Draft plannedLessons from building retrieval and GenAI systems for doctors, patients, operations, and compliance-heavy healthcare environments.
Draft plannedNotes on evidence, safety, fairness, and governance infrastructure for AI adoption across healthcare providers and public health settings.
Draft plannedI am open to research collaborations, healthcare AI discussions, responsible AI policy work, and conversations with teams building clinically useful AI systems.