As our lead software engineer, Colin Gagnon knows all about the complexities and opportunities of the Gazelle.ai system. His background in AI extends to over 10 years ago, and he’s putting all his knowledge into helping develop the platform with the entire Gazelle.ai team.

How long have you been working on the Gazelle.ai platform?

I joined the Gazelle team back in May after working in a few other early stage startups around Montreal, and having prior experience building products like this, I was immediately interested in the challenges the team would be tackling moving the technology forward and getting the product into market.

 

What’s your background in artificial intelligence?

I took an AI course at the Munich University of Applied Sciences in 2006, which, back then, was a purely theoretical exercise covering the history of AI going back to the 1950s and largely focussed on the limitations to implementing the various theories.  As a result, I have been closely following the advances and tools that have been created in recent years to allow developers to leverage these incredibly complex technologies and have been eagerly pursuing opportunities to work with more data, statistics, and machine learning.

 

From your perspective, what kind of background helps you get the job done?

In my experience, achieving a good result requires drawing on quite a few different disciplines that help in different ways. A background in software development makes learning the tools quicker and helps you write more efficient applications, whereas a background in statistics will help in working with large amounts of data and understanding important characteristics of your data. Likewise, an understanding of the business, its customers, and the market the company is competing in can help make good decisions about how to look at the data and what insights will be most valuable to the organization.

 

What are some of the biggest headaches (aside from doing interviews!) when it comes to the day-to-day work behind such a complex application?

In this type of application, there is always an ongoing focus on all things data.  Improving the quality of our data by detecting anomalies, removing duplicates, and keeping it all up to date and current. Adding new data from additional sources and matching to existing records to enrich the data that is available in the application to our users.  Creating new data sets from existing sources for use in training AI algorithms and validating the insights derived from them.

In addition to data-related things, allowing the data to be searched in many different ways and presented in the way that is most useful for the user is also an ongoing effort. We are constantly applying fixes and adding improvements to the application and listening to the feedback from our growing base of users is critical to this process.

 

Are AI and economic development strange bedfellows? You probably didn’t think of them together before you started working here…

Like anything that is really innovative, it is something that is really obvious in hindsight, but not something you would have considered beforehand.  Learning more about economic development every day, I can see more and more how you can use AI to focus the efforts of humans engaged in the evaluation of opportunities.