About Me
My background is in data science and applied machine learning research — including deep learning, reinforcement learning, convex optimization, and recommender systems / personalization. My primary interest is in the design of robust, scalable decision making systems. In my free time, I enjoy coding ML algorithms and web apps in Python and Julia.
I obtained my PhD in Computational Marketing from Columbia Business School (2022), co-advised by Prof. Asim Ansari and Prof. Kamel Jedidi. I hold an MS in Operations Research (MS&E) from Stanford University (2016) and a BS in International Business from Taras Shevchenko University (2012). While at Stanford, I worked as a research assistant at Stanford Intelligent Systems Laboratory (SISL) of Aeronautics and Astronautics Department, under Prof. Mykel J. Kochenderfer, doing research on deep reinforcement learning. Prior to that, in 2013-2014, I spent a year working for BlackRock Global Client Insights team.
Teaching
Books
Python, Deep Learning, and LLMs: A Crash Course for Complete Beginners
www.python2llms.org | Buy on Amazon
This book is a coding and machine learning bootcamp in textbook form. Put in the work, and soon you will know, in concrete detail, how to create a miniature mind inside a chunk of silicon using code. If that is not magic, what is?
Entrepreneur's Bad Advice: How to Get Rich
www.entrepreneursbadadvice.com | Buy on Amazon
The ultimate guide to becoming an oligarch is finally here!
Research Publications
Position: Enforced Amnesia as a Way to Mitigate the Potential Risk of Silent Suffering in the Conscious AI
ICML Poster | OpenReview | Project Page | .pdf | Video
This piece explores a philosophical hypothetical: if we are not sure whether a given system is sentient, could we mitigate the risk that it might be suffering by tampering with the system's memory? The work generated some heated discussion on Hacker News and received some video blog coverage.
A Megastudy on the Predictability of Personal Information From Facial Images: Disentangling Demographic and Non-Demographic Signals
Supplementary Information | Code & Data
(Older version is available on SSRN, titled "What Personal Information Can a Consumer Facial Image Reveal?")
At its release, the largest study of what personal information can be predicted from facial images — in the number of considered personal data variables. This project, at its various stages, has received energized book [1, 2] and blog coverage.
Reining in Long Consumer Questionnaires with Self-Supervised Deep Reinforcement Learning
This work develops an algorithm to get a question sequence that minimizes information loss when a consumer survey is cut short at an arbitrary point.
Scaling up Ranking under Constraints for Live Recommendations by Replacing Optimization with Prediction
Code | Research Brief | VentureBeat Piece
This work develops a way to approximately solve larger constrained ranking problems real-time, within 50 milliseconds, than previously reported.
Machine Learning Algorithms for Efficient Acquisition and Ethical Use of Personal Information in Decision Making
Advisors: Prof. Kamel Jedidi and Prof. Asim Ansari.
Pet Projects
Proof-of-Concept Web App for Matching People for Any Purpose
Side-project put together over a couple of weeks in Jan 2024.
Cases
Tutorials
Sample Size Determination
A tutorial on sample size determination – in marketing research and elsewhere – in a book chapter format.
Intro to R Programming
As prepared for edX Marketing Analytics course taught by Prof. Asim Ansari and Prof. Kamel Jedidi.
Logistic Regression for Churn Modeling and CLV Optimization
As prepared for edX Marketing Analytics course taught by Prof. Asim Ansari and Prof. Kamel Jedidi.
Other Writing
Misc.
The Car and the Pedestrian Experiment by Fons Trompenaars pdf
Stanford Marshmallow Experiment by Walter Mischel wiki pdf
Responsibility-for-Plant Experiment by Ellen J. Langer and Judith Rodin pdf
Basic Econometrics Lectures by Arne Hallam site
Gaussian Components by Peter Orbanz pdf
CS231n Convolutional Neural Networks Course Notes by Andrej Karpathy site
Information Cascades by David Easley and Jon Kleinberg pdf
Decision Making Under Uncertainty by Mykel J. Kochenderfer book
Philosophy of Civilization by Albert Schweitzer book
A Theory of Human Motivation by Abraham H. Maslow text
Eros and Civilization by Herbert Marcuse wiki book
Atlas Shrugged by Ayn Rand wiki book
The Book of Ecclesiastes wiki text
List of Action Verbs for Resumes pdf
Convex Optimization by Stephen Boyd and Lieven Vandenberghe pdf
Discrete Choice Methods with Simulation by Kenneth Train pdf
Harbingers of Failure by Eric Anderson et al. text
Big Five Inventory pdf
Reference Notes From Gyan's Research wiki
Free Will Experiments by Benjamin Libet pdf
The Mind's Best Trick by Daniel M. Wegner pdf
Conscious Mind by David Chalmers pdf
Breakdown of the Bicameral Mind by Julian Jaynes book
CS168: The Modern Algorithmic Toolbox site
Compressed Sensing Illustration in Python site
20 Simple Distinct Colors site
Power and Sample Size Analysis Notes by Charles DiMaggio site
Big Maс Index Generator by The Economist code
Top 50 matplotlib Visualizations site
Humans May Sense Earth's Magnetic Field by Kelly Servick pdf
A Classroom Demonstration of Gullibility by Bertram R. Forer pdf
Intuitive Explanation of Discrete Fourier Transform as Correlation Analysis site
Make Matplotlib Plot Fonts Look Good With LaTeX site
Hoeffding’s D Explainer site
Concentration bounds via optimization site
PyTorch internals site
How many products should you offer on your paywall? site
Installing Homebrew and Anaconda on a Mac site
The Cost of Differentiating Through Optimization site
Microtubule-Stabilizer Epothilone B Delays Anesthetic-Induced Unconsciousness in Rats site
Programmatically Send Push Notifications to Telegram from Python site
Differentiable Logic Cellular Automata site
Computational Genomics with R by Altuna Akalin et al. site
Avoidance of sun exposure is a risk factor for all-cause mortality blog pdf
Defeating Nondeterminism in LLM Inference site
A brief history of random numbers site