This blog is part of the Research Software Camp series.
When moving from finance to oceanography, I realised research software needs the same commercial-grade rigour as banking systems. In this blog, I share my journey from full-stack software engineer to Research Software Engineer at the National Oceanography Centre, and how running Software Carpentry workshops helps researchers build better, more sustainable code.
Why software rigour matters in research
I spent years in Peru's fast-paced financial sector as a back-end and front-end developer, and technical lead. At major institutions like Credit Bank of Peru, Interbank, and Bank of Commerce, I built high-performance, fault-tolerant systems. In banking, downtime or bugs have immediate financial consequences. Software best practices aren’t optional—they’re essential.
When I moved to the UK to study for my MSc in Advanced Data Science at Bangor University, I discovered a different world. Working as a Software Developer at Marine Centre Wales, I saw firsthand how essential modern software practices are to scientific research. Marine scientists deal with massive, complex spatial datasets such as high-resolution tidal models, and turning that data into scientific insight requires robust software tools.
I quickly learned that bringing industry standards into academic environments transforms how science is done. Combining clean code with clear visualisation tools saves months of manual effort and helps researchers spot crucial trends in ocean data far faster.
Building tools for marine science
Today, I’m a Research Software Engineer (RSE) in the Software Engineering and AI group at the National Oceanography Centre (NOC). My work focuses on building tools that turn complex oceanographic datasets into intuitive, actionable visual insights. We bring commercial-grade engineering principles to ocean science through two key projects:
- SPLASH dashboard: A tool designed to predict wave overtopping hazards and keep coastal communities safe.
- AtlantiS visualisation tool: A platform enabling users to explore and combine multilayered marine datasets seamlessly.
Developing these tools reinforced a critical truth: brilliant algorithms aren't enough if researchers can't easily interact with or trust the software behind them.
Delivering Software Carpentry training at NOC
Creating tools is only half the solution. To truly improve software sustainability, we must equip researchers with fundamental computational skills.
At NOC, I regularly organise, instruct and assist Software Carpentry workshops. During these hands-on sessions, we teach scientists and data managers essential tools like version control with Git, Intermediate Python, Shell and Web APIs.
Teaching researchers has been one of the most rewarding parts of my role. Seeing researchers gain confidence in writing reproducible, clean code means they can spend less time struggling with data formatting and more time driving scientific breakthroughs.
Why software sustainability is a shared responsibility
Transitioning from commercial banking to marine research showed me that software challenges are universal. By sharing technical expertise and running regular workshops, we bridge the gap between commercial engineering standards and scientific research. If you are an RSE or researcher, I strongly encourage you to host or attend computational skills workshops at your institution. Equipping early-career researchers and research scientists with proper software skills makes research faster, more transparent, and more reproducible.
Get involved
- Looking to elevate your computational skills? Learn how to get involved or run a workshop through the Software Carpentry page.
- Discover how the Institute for Research Software supports software practice across the UK research community by visiting the Institute's website.