Eating lots of pizza because your local outlet always seems to know when your tummy’s rumbling? MakerOps Inc.'s Co-Founder and Chief Data Scientist Thomas Atkins might be to blame. He applies contextual information and behavioural data to algorithms and uses them to improve the customer experience at companies including Domino’s and Sportsbet — and he’ll be helping MakerOps Inc.'s customers to do the same.
Data science and analytics can be used to create amazing change, and like many others, I get satisfaction from discovering new patterns and information. But it only delivers real value when a data model is productionised, or put in use as part of a larger automated system.
I’m very keen to see that data gets used and implemented to create positive change rather than getting dumped into reports and never being actioned.
At MakerOps Inc. I’ll be smoothing this process for our customers, whether it’s by interconnecting various disparate data sources or by building ready-to-deploy workflows for applied machine learning.
The need for improved customer experience is constantly increasing and innovations are becoming ‘the norm’ much more quickly now than ever before.
But businesses are struggling to keep up with the growing pace of innovation, particularly in data and analytics and the development of real-time customer experiences. They need new tools and processes to enable them to keep pace with customer expectations. That’s where our platform comes in.
On the product side my role is focused on building intelligence features into functions and workflows as part of the development of our platform and services.
On the advisory side I’m working directly with customers to help them develop their technology roadmap and map out their real-time customer experience opportunities.
In a startup you’re all doing everything, but one of the key areas I’m focusing on is evaluating growth opportunities based on the financial model for the business and the market we’re operating in.
I haven’t so this is really new to me and super exciting. I’ll definitely be relying on my fellow co-founders who do have startup experience to guide me.
The great thing is we get full control — we’re not going into a business that has an existing set of values. We’re in control of creating a culture and the business can be whatever we want it to be.
The exciting part for me is learning about new algorithms and technologies, thinking about where they can be applied, and watching how they influence people’s lives.
Data science has experienced a great deal of growth recently, both in academia and industry — but while there’s a lot of interest in it there’s also a lot of hype about it.
I worked on the development of a targeted SMS platform that would send out personalised SMS to customers at the time of day that they normally interacted with the brand. This resulted in engagement occurring much closer to the time of send, thus avoiding messages being lost in the SMS history.
I’ve also worked with applied statistical models — in one case, using them to predict equipment breakdowns so that planned maintenance could be carried out before failures occurred in operation. This involved developing a statistical fingerprint of a machine’s operation, then monitoring the real-time data from the machine and comparing it to the fingerprint to look for deviations.
I’ve always been interested in mathematics and computer science (my undergrad degree was a double degree with engineering and math/computer science) and I maintained this during my engineering days by building small tools to solve problems.
Then in the late 2000s I became involved with maintenance data at work and also discovered Kaggle competitions and this ignited my interest into a passion.
When I left Anglo American in 2014 to take up a role at Sportsbet I stepped outside engineering and began to expand my knowledge into new areas of data science.
You’d find a very confusing mix of cooking shows, science fiction, tech podcasts, and My Little Pony!
Device sharing, like in my home, creates an interesting problem for data scientists particularly around identity resolution and profile enrichment processes.
I’d like to see a better understanding of my context at a given time. For example, am I at work and interested in seeing new tech products or data science news? Or am I at home, relaxing and looking for recipes or a new sci-fi show to stream?