With a 19+ year career spanning enterprise technology, early cloud innovation, and leadership roles in some of the world's most influential tech companies, Steve Bryen brings a wealth of experience to Mesh-AI.
He will play a key role in driving growth across Technology at Mesh-AI and delivering innovative solutions to our customers.
We’re delighted to welcome him to Mesh-AI and understand his perspective on how data is being harnessed across the enterprise, and the biggest challenges facing large organisations right now.
Tell us a bit about your background in engineering. Where did you start and what are some of your biggest learnings?
I started my career in enterprise technology on the client side, working at BT and JP Morgan, focusing on large-scale infrastructure and application engineering. At BT, I developed network monitoring applications for major clients like VISA and KPMG.
In 2010, I moved into early cloud computing with a service provider, building orchestration software and working with platforms like CloudFoundry and early AWS services – supporting major events like Comic Relief’s Red Nose Day.
I then transitioned into Big Tech, joining VCE and later AWS, where I was one of the first technical hires in EMEA and eventually led Solutions Architecture for the UK Energy and Utilities sector.
After a stint as Head of Data & AI Architecture at Microsoft, I returned to AWS, supporting some of the largest and most complex customers globally through Developer Relations and Strategic Accounts roles.
More recently, I’ve held customer-side leadership positions, serving as Distinguished Engineer and Director of Engineering at Showpad, and VP of Engineering at Oxa, a leading UK AI startup focused on self-driving vehicle technology.
From a data and engineering point of view, how do you see the state of data in the enterprise right now? What are some of the biggest areas enterprises need to work on?
We’re at a really exciting juncture for many enterprises. The hype generated by AI is driving investment in technology and most are realising that to capitalise on the opportunities at hand, you need to invest in how you manage and access your data.
Despite how enterprises are prioritising data, there are three major challenge areas.
1) Firstly, most are finding that quality is lacking and the answer is in their data infrastructure. This is one of the biggest challenges organisations need to solve. Large organisations have data at a vast scale and level of complexity, yet data quality continues to be a huge challenge. As teams experiment building AI solutions or adopting them off the shelf, inaccuracies or incomplete data sets emerge as a stumbling block to innovation. This stifles adoption and creates a barrier to AI succeeding.
2) Businesses need the right skills in house to build this critical technology infrastructure. In a competitive talent market, it’s not always possible to hire your way out of the problem – especially if you need to build a data science or engineering team from scratch. Choosing the right partner can help embed familiarity so you’re innovating and bringing people on the journey with you at the same time.
3) Lastly I’ve seen enterprises struggle to operationalise AI. Building demos and proofs-of-concept helps you prove the value of deploying AI in your organisation, but moving beyond this stage proves tricky for many. Today’s data is emerging from an ever-growing number of sources and at unprecedented volumes, making it harder to manage and trust.
To succeed, AI systems must be treated like any other enterprise technology - maintained, monitored, and scaled. This includes tasks like retraining models as new data becomes available, tracking performance drift, scaling infrastructure, and enforcing guardrails such as profanity or abuse detection in prompts. An ML/AIOps approach helps organisations manage the full lifecycle of their AI applications, ensuring they can move from experimentation to stable, scalable production systems.
We hear a lot about AI opportunities, particularly with new technologies like Agentic AI. But how much work do most enterprises need to invest in their data foundation still?
There’s a bit of a 90/10 rule when it comes to investing in AI. In order to get the most of AI and ensure long-term growth, this is how you should split your attention and resources between data and AI.
Hype in the public eye has driven boards to question internally what their businesses are doing about AI. Leaders have to respond, and this often results in AI solutions of limited scope, primarily targeted at cost savings or efficiency gains, instead of the long-term value that’s actually possible.
I feel this is because they still need to develop the foundational technical maturity necessary to adopt transformational solutions. As a baseline, this means storing vast data at scale, making it discoverable, queryable and usable.
New technologies such as Agentic AI hold much promise, and we’re already pioneering use cases for some of our customers who have consistently invested in maintaining their data infrastructure. It’s these enterprises who are ahead of their competitors and setting the standard when it comes to technological transformation.
For enterprise clients, what’s the biggest problem that we are uniquely positioned to solve?
It’s a great time to be joining Mesh-AI. I’m excited by the transformative solutions we’re delivering for enterprise organisations and the impact we’re making, be that bringing an AI content curation system into production in just 14 weeks or protecting the UK’s critical infrastructure.
There is a significant opportunity to ensure data is managed well and in a good place so enterprises can make the most of AI, giving our customers one eye on now and one eye on the future. Mesh-AI is uniquely positioned to help our customers with both.
From an engineering perspective, we’re building the platforms and solutions to put data in the hands of those who need it most, so they can innovate and drive their companies forward. By building the foundations and then capitalising on what’s next, our customers can lead in their own domains, and leapfrog their competition.