Nicolas Kruchten

Nicolas Kruchten
builds data visualization tools
at Plotly in Montreal, Canada.

Peeking Into the Black Box, Parts 1-5

Peeking Into the Black Box, Parts 1-5

Between 2011 and 2013 I wrote a popular 5-part series of articles about Datacratic's real-time bidding algorithms, and I've collected them together here for easier reading.

Full post »


JS-Montreal: PivotTable.js


Slides from a talk I gave at JS-Montreal about PivotTable.js

Full post »


PivotTable.js

PivotTable.js

When I wear my 'data scientist hat', one of the tools I reach for most often is a pivot table. When I wanted to build a web-based tool that included a pivot table, I didn't find any Javascript implementations that made sense or didn't have crazy assumptions built-in, so I rolled my own in CoffeeScript, as a jQuery plugin.

It's now up on GitHub under an MIT license with some nice examples. I hope people find it useful!

If you work with data and you don't know what a pivot table is, I encourage you to learn about them, because they are very useful for quick'n'dirty data analysis. My web-based implementation is a decent learning tool but there are other, much-better implementations, such as in Microsoft Excel (although since Office 2003 they've made some changes that were not for the better) and AquaDataStudio.

I posted this on Hacker News and got some nice comments!

Full post »


Kernel Density Estimation and Analysis Tool

Kernel Density Estimation and Analysis Tool

In 2005, I was contracted to create a program to support research into the application of a statistical technique called Kernel Density Estimation to the study of global poverty. The result of this contract (which I worked on with my friend and occasional colleague David de Koning) is the Kernel Density Estimation and Analysis tool which I have just released on Github under an open-source license.

The research (which, to be clear, wasn't done by me) resulted in a very interesting paper called Kernel Density Estimation Based on Grouped Data: The Case of Poverty Assessment.

Full post »


Galapagos

Galapagos

In 2003, I wrote a neat and powerful piece of software called Galapagos for my 4th-year undergraduate thesis (download PDF). It was a framework for the development of advanced (i.e. distributed, parallel and/or hybrid) evolutionary algorithms, applicable to a wide range of computational challenging optimization problems. I applied it to a variety of transportation-related problems at the University of Toronto.

Full post »


Rubicon Tech Talk: The Algorithms Automating Advertising


I was invited to speak on a panel at a Rubicon Project product launch, and this is the video of the event.

Full post »


Volunteer Database for Santropol Roulant

Volunteer Database for Santropol Roulant

There’s an organization in Montreal I think is awesome called Santropol Roulant which, among other things, has a meals-on-wheels operation. They have hundreds of volunteers and wanted to upgrade the system they used to store their volunteer information, so I helped them out, and I’ve open-sourced the results, in case any other non-profit wants a very simple volunteer-list management system.

Full post »


My QR-Code Business Card

My QR-Code Business Card

This is the machine-readable back of my new nerdy QR-code business card!

Full post »


Real Time Bidding, Characterized


There doesn't appear to be a good Wikipedia entry for RTB for me to link at the moment, when I want to blog about it so I'll draft my own explanation here. (Edit: there is an entry now, but I like my characterization better!) Keep in mind while reading this that I'm looking at RTB as a software engineer with an interest in economics, rather than as an ad industry veteran!

Full post »


Using make to Orchestrate Machine Learning Tasks

Using make to Orchestrate Machine Learning Tasks

One of the things we do at Datacratic is to use machine learning algorithms to optimize real-time bidding (RTB) policies for online display advertising. This means we train software models to predict, for example, the cost and the value of showing a given ad impression, and we then incorporate these prediction models into systems which make informed bidding decisions on behalf of our clients to show their ads to their potential customers.

Full post »



© Nicolas Kruchten 2010-2021