Director of technology at Liberty IT, Stuart Greenlees, shares some of his top learnings from delving into the world of AI and his top advice for those starting out.

Click here to view the full AI and Analytics Week series.

Stuart Greenlees has worked at Liberty IT for about 20 years and in that time has gained a breadth of experience working across the entire organisation. In his current role as director of technology, he focuses on innovation. “I joined the company because I was looking for more complex challenges to work on and I haven’t been disappointed,” he told SiliconRepublic.com.

Greenlees is particularly interested in how emerging technologies and professional practices can be utilised to solve emerging business problems for the tech arm’s parent company, Liberty Mutual Insurance, which he said insures “everything from houses to airplanes and horses, so there is lots of variety in the types of things you can work on which keeps it interesting.”

‘You need to be able to communicate to understand the problem you are going to solve’

What first stirred your interest in a career in AI and analytics?

I first got interested in AI and analytics about seven years ago when I took on the role of senior portfolio architect. What I quickly discovered was that software engineering alone was not enough to solve some of Liberty Mutual Insurance’s most complex business problems.

Since then, I’ve been heavily involved with AI. Not only have we used commodity AI for things like cognitive experiences and document intelligence, but we have also built up a data science team who build custom predictive, computer vision and NLP models and a machine learning engineering discipline with a focus on ML Ops.

For the last year I have been heavily focused on generative AI and how Liberty Mutual Insurance as an enterprise can take advantage of this powerful technology.

What were the biggest surprises you encountered on your career path?

In AI and analytics there have been lots of surprises and challenges which have kept it both challenging and interesting.

One of the biggest surprises is the number of technical challenges around an AI solution beyond building the actual ML model. For example, it’s often difficult to get the data you need. Once you have that, it’s difficult to get it labelled correctly.

There are challenges with tracking your experiments along with your models and associated data and feature engineering code. Once you have a model, there are a host of other challenges with regards to things like operationalising the model, monitoring its performance, watching for data drift and refitting the model.

In Liberty, we have spent the last few years building our ML Ops practices and tooling to tackle these challenges with the goal of reducing both the cycle time it takes to get models into production and the amount of time our data science teams need to spend on refitting models rather than building net new models to add to our portfolio of models in production.

Was there any one person who was particularly influential as your career developed?

Not just one, so many! I have had several mentors who helped me build my technical and professional skills to whom I’m incredibly grateful. Also being a mentor is extremely important to me.

What do you enjoy most about your job?

As someone who’s really interested in technology, I really enjoy learning about emerging technology and then applying that technology in the real world to solve problems that have an impact for our business and customers. As much as I enjoy experimenting with technology and proving out ideas with proof of concepts, neither of those things really count until the solution is in production making a difference.

The other thing I really enjoy is working with other people, particularly when you can help someone use their passion for technology to progress in their career.

What are some of your key learnings from working with AI?

I think coming from an engineering background I am used to the engineering role expanding to take on more and more elements including things like testing and security for example. So, one of the biggest learnings for me as I have become more involved in AI and analytics is the need for a good foundation in mathematics and statistics. For more open problems, particularly when there is no clear solution, data analysis and statistical evaluation are critical to realise value for the business.

What can people expect from career progression in the AI and analytics industry?

With AI and analytics being such a key area within the industry there are so many opportunities for career progression. In Liberty IT we have a learn-it-all mindset and helping people progress in their career is a key part of our culture. Liberty IT also helps engineers with a strong scientific background to make the switch to data science and has helped our data scientists develop their ML Ops skills.

What advice would you give to those considering a career in AI and analytics?

When starting out in a technical discipline it is important to really focus on the technical aspects of that discipline. For AI, that might mean a strong foundation in statistics, programming in Python or R and machine learning concepts.

It’s also incredibly important to develop your communication skills as solutions are rarely created in isolation and you need to be able to communicate to understand the problem you are going to solve.

I’d also recommend finding someone with more experience than you who can mentor you and give you technical and career advice and hold you accountable to progress.

Finally, I’d also really encourage remaining curious and focusing on a continuous learning mindset. Things are changing constantly in this domain and spotting new skills, tools or practices that you can use to help you solve problems will really help you improve your professional practice and progress in your career.

Find out how emerging tech trends are transforming tomorrow with our new podcast, Future Human: The Series. Listen now on Spotify, on Apple or wherever you get your podcasts.

Source link