Hey! My name is Jonathan Strong, welcome to my personal website.

The articles I publish here are predominantly about software, especially the Rust programming language, with the occasional foray into other interests.


Intrepid reporter Jonathan Strong holds politicians' feet to the fire

I had to wear a suit every day. Can you imagine?

Ten years ago, I was a political reporter. Now I write computer code for Keyrock, a cryptocurrency market maker.

This unusual path began in high school when I learned C++, launching a deep interest in coding. In a more immediate sense, it began in the press gallery of the U.S. Capitol where I first started using computers to analyze the politicians I was writing about.

From an initial database for tracking votes, campaign spending, and other data, a startup was born.

The plan was to forecast the outcome of congressional votes using a proprietary combination of subject-matter expertise and cutting-edge tech, selling our predictions to hedge funds who could use the informational edge to anticipate market-moving political news.

At one point, we even scored an Economist profile:

[Strong] has spent three years building Legis, an algorithm powered by vast quantities of data and a neural network (a computer system modelled on the human brain), which predicts the outcome of congressional votes. Each of the 44 votes it has forecast so far has been correct. Last year a hedge fund (which does not want to be named) began trading derivatives using its predictions

The Economist, Feb. 28, 2018

Alas, it was not to be. Like most startups, we ran out of money. Nevertheless, it was an amazing journey, both in terms of life experience, and also as a unique learning opportunity forged from coding at every level of the stack, working alongside some incredible data scientists and building a production machine learning pipeline from scratch.

The work also kindled another deep interest, in financial markets (and timeseries forecasting, the emotional roller-coaster of trading, event-based, realtime systems, the asymmetric "information topology" of a market, and the amazing amount of stuff that can happen on a computer in an imperceptible, infinitesimal span of time ... the list goes on and on).

cool hft chart

Not gonna lie, I thought this btc/usd chart was pretty cool when I made it

Another thing I realized along the way is, a reporter's skill is learning new topics quickly: the stakes are huge (and public), the deadlines brutal. That made it easier to "rftm" and also ask the right questions.

My work on Legis, and some incredible times running arbitrage bots during the wild Bitcoin rally of 2018 (nothing teaches you how to build resilient systems like letting a computer trade your money while you sleep) led me to my current day job at a prop trading firm. Work is mostly Python and Ruby, but my preferred language is Rust, an incredible tool for building programs that are not only blazing fast, but incredibly robust.


I live in Northern Virginia, just outside of Washington, D.C., with my amazing wife Anna Beth and our three kids, ages ten, eight, and six. We spend a lot of time hiking and exploring outdoors, and I'm hugely into fishing, especially fly fishing for Smallmouth Bass on the Potomac.

I've been a Washington Capitals fan since I was a boy and their Stanley Cup victory in 2018 was a lifelong dream. I carry some pretty heavy anti-Pittsburgh Penguins baggage from the repeated playoff losses we suffered at their hands in the 1990s.

I graduated from Wheaton College, IL, where I studied political science (I did fit in some freelance programming work here and there).

After college, I climbed almost every mountain in the Tetons. Every June I wonder how it's possible I haven't been back.

mmm... pittsburgh penguin blood

On my birthday, my wife surprised me with a Pittsburgh Penguins piƱata

at the very top of the south teton

Me, smiling at the summit of South Teton (12,514)