Isaac Dimitrovsky's stuff!
"better living through awesomeness"

Machine Learning

Guru Predictor Have you ever wondered what stocks your favorite investors like Warren Buffett or Carl Icahn will buy next? Guru Predictor uses machine learning to predict the future stock buys of major investors based on their current holdings.
How to
Screw Up Stats
An amusing slideshow on the many ways to screw up statistical prediction, including a miscellany of instructive case histories.
How to Set Up
a Cheap ML Rig
While cloud ML environments such as Colab or Sagemaker are cool, it can also be nice to have your own desktop Linux ML rig - no worries about running into arbitrary limits or unexpected charges. A collection of tips and resources to make it happen for cheap.
St. Bernard
or Black Swan?
A Kaggle notebook demo in which we confuse an image classifying neural network into thinking a St. Bernard is a black swan (or whatever else we want).

History and Economics

News from 1930 A day-by-day news summary from the start of the Great Depression (pre-Roosevelt) - favorably reviewed in the Financial Times Alphaville, the New York Times Economix, Paul Krugman, The Atlantic, and the Wall Street Journal.
Tall Tulip Tales A few memorable stories from the 17th-Century Dutch Tulip Madness period. Plus some totally unrelated observations about Bitcoin. Written June, 2018.
Vas ist das
If you are secretly unsure about just what the hell Bitcoin actually is but find yourself ashamed to ask, this essay is just for you! With 10 minutes of mental effort, you'll understand Bitcoin better than most of the people talking about it on TV. Written July, 2018.

Stock Writeups

I wrote these for SeekingAlpha, a popular financial website.

IEH Corp. IEH Corp. (IEHC) - A Good Business at a Cheap Price, Undergoing a Promising Generational Transition. Written June 5, 2017.
Alloy Steel
Alloy Steel International (AYSI) - For the Loser Now Will be Later to Win, For the Catalysts, They Are a-Catalyzin'. Written June 19, 2017.
Rubicon Technology (RBCN) - A Promising Net-Net Cash-Box Situation. Written July 18, 2018.

Professional Information

Summary: Creative software engineer with long-term record of quickly developing reliable high-performance applications for clients ranging from hedge funds to major museums. Expert at algorithm and heuristic development and optimization. Working knowledge of financial security analysis, machine learning, and computational chemistry / biochemistry. Accomplished writer and presenter, skilled at explaining technical concepts clearly. Adept at learning new technologies.

LinkedIn My LinkedIn profile.
Resume My latest resume.
Email me Send me email at