Rhys Cadogan
Founder and lead engineer
Eight years in data science across fintech and logistics. Rhys handles project scoping, model architecture decisions and client relationships.
Who we are, why we started and how we think about the work.
Accrue AI Value started in 2021 when our founder, Rhys Cadogan, left a data-science role at a Cardiff fintech to help smaller companies access the same kind of machine-learning tools that large corporations take for granted. The first client was a family-run food distributor in Swansea that needed demand forecasts to reduce waste. That project cut spoilage by 18% in the first two months.
Word spread. By mid-2022 the team had grown to four, and we moved into a small office on Bergnaum Ridge in North Feil. We chose Wales deliberately: lower overheads mean we can price projects fairly without cutting corners on engineering quality.
Today we are a team of seven. We have delivered 84 projects across retail, logistics, manufacturing, healthcare administration and professional services. Every project follows the same principle: build something useful, hand it over properly, and make sure it keeps working after we leave.
Four ideas that shape how we run projects and make decisions.
If your data cannot support the outcome you want, we will tell you before you spend money. We have talked clients out of projects when a simpler spreadsheet formula solved the problem. That honesty is why 97% of our clients come back for a second engagement.
We aim for a working prototype within two to three weeks, not two to three months. Fast feedback loops mean you can test assumptions early, change direction cheaply and avoid the sunk-cost trap of long waterfall projects.
We never use client data to train models for other clients. Every dataset is stored in encrypted, region-locked infrastructure and deleted once the project wraps up. We sign NDAs before any data changes hands.
A model nobody on your team understands is a liability. We write plain-language documentation, run training sessions and pair-program with your developers during handover. The goal is that you can maintain and extend the system without us.
Small enough that you will know everyone by name. Experienced enough to deliver on time.
Founder and lead engineer
Eight years in data science across fintech and logistics. Rhys handles project scoping, model architecture decisions and client relationships.
Machine-learning engineer
Priya builds and deploys models in production. Before joining, she worked on recommendation systems at an e-commerce company in Bristol processing 2 million events per day.
Data engineer
Tom designs the pipelines that feed models with clean, timely data. He spent six years at an NHS trust building ETL systems for patient-flow analytics before moving to the private sector.
A short timeline of milestones since we started.
Rhys registers Accrue AI Value as a limited company in Wales. First client engagement: demand forecasting for a food distributor in Swansea.
Completed ten projects. Hired Priya as the first full-time ML engineer after she freelanced on two early contracts.
Moved into the Bergnaum Ridge office. Tom joined to handle growing data-engineering demands as clients started requesting pipeline work alongside model training.
Reached 40 completed projects. Launched the Enterprise retainer plan after three clients asked for ongoing embedded support.
Team grew to seven. Opened a second focus area in computer vision after delivering quality-inspection systems to two manufacturing clients.
84 projects delivered. Working with clients in 12 industries across the UK and Ireland. Still based in North Feil, still answering the phone ourselves.
Drop us a line and we will set up a discovery call within 48 hours.
5 Bergnaum Ridge, North Feil, Wales, GA41 8DZ, United Kingdom