15 minutes with: Neil Montgomery

Based in PA’s Belfast office, Neil Montgomery is working on an artificial intelligence (AI) system for knowledge management in a central government department. Coming from a startup background, he’s bringing ingenuity to life through experimentation, asking the right questions, and working directly with users to maximise positive human impact.

What is your background and how did you come to PA?

I’m a software engineer by trade and started out with a company who built systems for the government sector. I then worked for several large multi-nationals building a variety of applications in pharma, insurance, and ecommerce. Most recently, I spent seven years at a manufacturing start-up. We built a platform that enabled manufacturing companies to convert CAD (computer aided design) models into web compatible 3D models, and provided an ecommerce solution to help them quickly identify and order new parts. I was ready for a change, so I joined PA in September 2022. It was the variety of work that drew me to PA, and the inclusive community.

What makes working at PA different?

There are great opportunities to work on brilliant stuff. I'm currently working on AI innovation for a government department.

Our approach is to start with the problem. What keeps clients up at night, and can we solve these problems with AI? A lot of the time, technology is forced on organisations, rather than looking at the problem and then finding the right solution. There’s also been push-back against AI because of the narrative that it’s here to take everyone’s jobs. Approaching it from the end users’ side first is absolutely vital to our success, because you get to much better solutions and they are bought in as they know you’re trying to solve their problem.

Another thing that sets PA apart is the broad spectrum of people we can bring to assignments. One of my colleagues is a PA business strategist. We’ve been working as a double act – she comes from the business problem side, I come from the tech side, and we meet in the middle to get to the root of helping people.

If you had to describe what you do to a person you’d just met, how would you explain it?

I’m building an AI system that will be used by a government department to search through all their knowledge bases, helping them in their intense environment. We're training the system on all the department’s documents so officials can ask plain text questions and get answers. The system references the document and the specific page to show where the answer is from so officials can verify it. It's about making their job simpler and reducing the time they spend searching through the internet and SharePoint, making it easier to bring real value to their organisation.

What’s the role of ingenuity in the work that you do?

I’m working right at the cutting edge of generative AI. It's brand new. It’s taken a lot of trial and error to get to where we are now, to see what worked and learn from that. New techniques come out every day, so there's a lot of evaluation.

What inspires you in your day-to-day work?

The wave of AI is taking everything by storm. Every day, there’s something new that can be applied to new use cases. I came from the startup world, where we focused on differentiating ourselves through solving problems that matter. That's something I love and really enjoy – solving problems and looking at what technologies are out there to do it.

What developments in your industry do you find most interesting?

My thoughts are that new large language models (LLMs) will create a new user interface using text and speech. You will no longer have to find 12 buttons and memorise that series, and you don't have to use specific keywords or technical jargon when searching. You will be able to just say “I need to book my car in for an MOT” and it will do it for you. It knows how far away the MOT centre is and can book the MOT based on your calendar. These applications will make life easier and simpler for people, and take away a lot of pain and mundane tasks. It’s the human impact of those developments that’s really interesting.

One of my favourite things to do is to ask AI to explain a concept or a system to me like I'm five. It allows you to take complex technical language – for me, it’s often business strategy terms – and convert it into something easy to understand and then apply.

Of the assignments you've worked on so far, which are you most proud of?

I’m most proud of the project we’re working on in a central government department, mainly for how quickly we've moved in such a short period of time. It’s shown that if you embrace and apply technologies, you can move fast.

We're doing our first proof of value right now to demonstrate the business case and unlock the next round of funding.

This project is very personal because you work with end users day in, day out. We see the technology used in real-world settings, then chat with users and make it better. You see the smiles on people's faces when they realise it actually solves a problem. That's part of why this project is such a success.

What are your goals professionally?

One of the things I've always loved is mentoring and coaching other people, to share knowledge about AI and LLMs and how to build these systems. Going into the technical director pathway would be the best of both worlds for me, because it lets you build relationships, teach people, and help grow the business, but you can still get your hands dirty writing code and doing the fun technical stuff.

What advice would you have for somebody who wanted to work on AI innovation?

My advice is to just get started. Ask to be part of an upcoming AI project, propose one, or simply start dabbling with it in your spare time. There are lots of good tutorials and courses available, and there’s no better time to get involved in a new technology than at the start where you can learn the fundamental building blocks at each stage as it grows and evolves.

Technology is one of those areas where if you get too comfortable, you'll probably get left behind. The pace of change is just so fast. 

You’re based in PA’s Belfast office. What’s the vibe like?

Most people who visit the Belfast office try to find an excuse to come back as soon as possible. It's predominantly a digital hub, which I've enjoyed a lot because everyone is working on cool projects with different tech stacks. There are around 100 staff in the Belfast office and I think 85 of us are in digital, if not more. You can't turn around without bumping into a technologist.

You learn something new all the time, and you can ask questions of anyone in the office about anything they’re working on. It's very supportive and extremely interesting. The office is continuing to grow, and in the right way, furthering the supportive and inclusive environment. Everyone has the right outlook and optimism for what we're trying to do.

What digital projects are your fellow Belfasters working on?

Take any type of technology project kicking around in any corporation, big or small, and there's a good chance we have someone working on something very similar. We have projects across the whole spectrum, including managing massive legacy applications, mobile app development, cloud infrastructure projects, to AI… We have testers and Quality Assurance analysts working on government projects serving millions of users. The level of infrastructure that our DevOps engineers are creating around that is impressive. We have engineers and architects working across all the major cloud providers: GCP, AWS, and Azure.

What are you looking forward to right now?

I’m working on an eight-week project to build a knowledge assistant, and I’m really excited to get it to real users. I can’t wait to see them use it in 50 different ways I've never thought of, then learn and iterate from that to create new features and functionalities that solves problems for them. It's so novel and different to anything else we've done up to now, so I'm excited to see it in the hands of real live users.