I build probabilistic models that reveal the hidden processes behind health inequalities — turning noisy, incomplete data into interpretable insights that matter.
Documents
Public versions of my current CV and research summary are available below.
Learning interpretable latent structure from noisy, heterogeneous, and spatially evolving data — to understand and address unequal health and developmental outcomes.
My doctoral research develops probabilistic, spatial-temporal, and machine-learning models that infer unobserved processes — access, engagement, risk, developmental vulnerability — from large-scale observational data across low and middle-income countries.
A defining feature of my work is treating latent variable modeling as a tool for substantive inference, not merely dimensionality reduction. Observed outcomes are modeled as consequences of underlying processes shaped by social, structural, and contextual factors.
Going forward, I aim to extend this into deep latent dynamical models that infer meaningful states from irregular, multimodal data streams — preserving interpretability and uncertainty-awareness throughout.
Five projects are listed below. Each card opens a dedicated page with explicit details on problem statement, motivation, logic, modeling, and mathematical approach.
Multi-country study on how socioeconomic factors shape vaccine trust and behavior. Latent variable models decompose demand from access barriers across diverse national contexts.
ML models identifying determinants of developmental variation in rural India, including active participation in contextual adaptation of ECDI survey instruments for village settings.
Predictive models for early identification of zero-dose children across the immunisation cascade, explicitly separating model performance from substantive interpretation.
Sub-national Bayesian spatial analysis identifying vaccination coldspots in India using SPDE via INLA, modeling spatial dependence, regional heterogeneity, and measurement uncertainty.
Extension of Zero-Dose modeling into a fresh retraining framework for ZD, MCV dropout, and DPT dropout with explicit national-vs-regional NN behavior, LR comparison, and deep interpretability analysis.
Doctoral research on latent variable modeling for health equity. TA for Advanced Machine Learning, Bayesian Modelling, and Machine Learning. TCS Research Scholar Fellow (2023–2027).
Mitacs Globalink Research Award. International research collaboration on immunisation modeling and health equity.
Selected as a YHP Changemaker Scholar, representing global health research and advocacy.
International dialogue on responsible AI governance in peace and security contexts.
Collaborative data science challenge on multimorbidity and long-term conditions in England using NHS open data.
Taught undergraduate AI and data science courses.
Risk and financial analytics in enterprise consulting, delivering data-driven solutions for large-scale clients.
Live streaming dashboards and statistical pipelines for the 2019 Lok Sabha General Election using ARIMA, LSTM, and Tableau.
Research on Bhavantar Bhugtan Yojana and mobile banking adoption using statistical modeling.
2023 – 2027 · Competitive fellowship from Tata Consultancy Services for doctoral research excellence
2024 – 2025 · International research award for collaboration at Université de Montréal, Canada
2025 · Young Health Programme — delegate to the 30th AFS Youth Assembly
Professional certifications, training programmes, and credentials earned alongside doctoral research. Click any card to open the certificate.
Download my complete CV — publications, projects, teaching, and more.
08 — Let's Connect
Actively exploring postdoctoral and research scientist positions in Bayesian machine learning, health data science, and computational social science.