May 2025 – Present · 1 yr

May 2023 – Apr 2025 · 1 yr 11 mos

• Advisors: Elizabeth Stuart and Kara Rudolph
• Collaborators: Avi Feller, Melody Huang, Rachael Ross, David Arbour, Caleb Miles, Trang Nguyen
• Project/Papers:
• 1. Characterizing Underrepresented Populations
• 2. Data Fusion with Disparate Outcome Measures
• 3. Regularizing Extrapolation in Causal Inference
• 4. Toward Generalizing Trials Inferences to Target Population

May 2025 – Present · 1 yr

• At Danish Centre for Health Economics (DaCHE) and Department of Public Health (IST)

Apr 2025 – Present · 1 yr 1 mo

May 2022 – Aug 2022 · 3 mos · New York, New York, United States

• Team: Causality and Privacy Research Group
• Project: Inferring Network Interference from User Randomized Experiments

May 2021 – Aug 2021 · 3 mos · Seattle, Washington, United States

• https://github.com/amzn/credence-to-causal-estimation/tree/main/credence-v2

Jun 2020 – Sep 2020 · 3 mos

• Credence to Causal Estimates: Designed a 'validation of causal estimation methods' framework. The framework learns the parameters of a simulator for generating data that imitates the dynamics of real world data of interest. It generates a synthetic dataset with known ground truth causal effect using the learned simulator to validate the performance of causal estimation methods based on their ability to recover true treatment effects.

Aug 2018 – May 2023 · 4 yrs 9 mos

• Advisors: Cynthia Rudin, Alexander Volfovsky, Sudeepa Roy;
• Thesis: Interpretable Causal Inference for High-stakes Decisions;
• Causal inference methods are widely used in various fields to assist in making critical decisions. However, these methods heavily rely on strong assumptions that may not always hold in high-stakes situations. When incorrect assumptions are made, the resulting inferences can lead to suboptimal decisions with severe penalties and long-lasting consequences. Unlike prediction or machine learning approaches, evaluating the performance of causal methods solely based on observed data is particularly challenging because we lack the ground truth causal effects for all units.
• To address this issue, my research proposes three frameworks to validate causal inference methods:
• (i) The first approach involves auditing the estimation procedure by a domain expert. This allows experts in the specific field to critically evaluate the procedure and assess its validity based on their knowledge and expertise.
• (ii) The second approach focuses on studying the performance of causal methods using synthetic data. By creating artificial data that mimics real-world scenarios, researchers can assess how well the methods perform and identify any limitations or biases.
• (iii) The third approach utilizes placebo tests to detect biases. Placebo tests involve applying the causal inference method to data where no true causal effect exists. By observing the results in these placebo scenarios, researchers can identify any biases or confounding factors that may affect the estimation procedure.
• By employing these validation frameworks, decision-makers gain the ability to assess the validity of the estimation procedure and make informed judgments by critically considering the underlying assumptions.

Jan 2017 – Aug 2017 · 7 mos

• Public Finance and Fiscal Policy, under guidance of Professor Juan Carlos Suarez Serrato, Duke University

Aug 2016 – May 2019 · 2 yrs 9 mos

• 6) Graduate Teaching Assistant for COMPSCI 671 - Machine Learning. Spring'19
• 5) Graduate Teaching Assistant for COMPSCI 590 - Computational Microeconomics. Fall'18
• 4) Graduate Teaching Assistant for COMPSCI 223 - Computational Microeconomics. Spring'18
• 3) Graduate Teaching Assistant for COMPSCI 230 - Discrete Mathematics. Fall'17
• 2) Graduate Teaching Assistant for COMPSCI 230 - Discrete Mathematics. Spring'17
• 1) Graduate Teaching Assistant for COMPSCI 201 - Data-structures and Algorithms. Fall'16

Jun 2017 – Jul 2017 · 1 mo · Washington D.C. Metro Area

• Center for International Development and Governance (IDG) -
• 1. Public Transport and Rental Market in Lahore:
• Performed causal analysis on impact of metro-bus service on ridership patterns across occupations, income groups and genders. Results showed metro-bus to be preferred mode for low income servicemen. Studied the variation in housing rents 3 years before and after metro-bus service's induction. Deduced that, in Lahore, expenses on amenities dictate rents more than access to metrobus station.
• 2. Women Empowerment and Labor force participation:
• Analyzed Tanzania, Senegal, Nigeria and Madagascar's household survey data to understand the effect of women's labor force participation with decision making power. Results showed the positive correlation but of varying degree across cultures. Studied the impact of shared economy initiatives in international tourism on women empowerment highlighting the lack of empirical evidence.

Jan 2017 – Apr 2017 · 3 mos · Raleigh-Durham, North Carolina Area

• North Carolina Voucher Program evaluation:
• Led a 6 member team for narrative construction from interview data of 60+ school administrations and parents on K-12 private school opportunity scholarship program. Digitized 500+ private school data to enable empirical study on effectiveness and equitability of the voucher program.

Jul 2015 – May 2016 · 10 mos · New Delhi Area, India

• Social network data analysis for law enforcement-
• Developed computational method for suspect identification from twitter network based on characteristic matching, location mining, tweet analysis & network's graph structure, to enable the law enforcement agencies to track the activities of the suspect based on his social-network updates.

Jul 2014 – Jun 2015 · 11 mos · New Delhi Area, India

• Ocean health prediction by Image Analysis using Convolutional Neural Networks:
• Developed machine learning algorithm aided by image processing techniques to automate the plankton image identification from underwater image sensory by designing a phylogeny inspired hierarchically stacked fine-tuned classifier model.

May 2014 – Jul 2014 · 2 mos · India

• Audio Video Bridging team developing product for ESPN.
• Implemented PTP and MSRP protocol features for Audio Video Bridges.

Aug 2013 – Dec 2013 · 4 mos · France

• Machine Learning Project in Neurosys/Cortex team on Neuroscience.
• Developed a Clustering Algorithm to aid in finding the periodicity of brain waves. Comparative analysis of existing Clustering Algorithms on workability on Recurrence Quantitative Analysis Plots

May 2013 – Jul 2013 · 2 mos · New Delhi Area, India

• We have carried out an analysis on 500 bacterial genomes and found that the de-facto GC skew method could predict the replication origin site only for 376 genomes. We also found that the auto-correlation and cross-correlation based methods have a similar prediction performance. In this paper, we propose a new measure called correlated entropy measure (CEM) which is able to predict the replication origin of all these 500 bacterial genomes. The proposed measure is context sensitive and thus a promising tool to identify functional sites. The process of identifying replication origins from the output of CEM and other methods has been automated to analyze a large number of genomes in a faster manner. We have also explored the applicability of SVM based classification of the workability of each of these methods on all the 500 bacterial genomes based on its length and GC content.