03 · Analytical product · Personal project
RunLab
A live product translating established running-performance methodology into practical pacing, race-equivalence and training tools. Designed and built end to end using AI-assisted development.
Product design · Performance modelling · AI-assisted build
What this is
Running produces plenty of data. The harder problem is turning established performance methodology into useful decisions about pacing, equivalent performances and training intensity.
RunLab is a published product exploring how analytical models become simple, practical tools rather than remaining abstract calculations — the same discipline as good management reporting, applied to a different subject.
What was built
Race pacing
Track pacing for 800m, 1500m and 3000m with optimal and even approaches, plus road pacing for 5K, 10K, half marathon and marathon — a target turned into checkpoint splits that can be used on the day.
Performance modelling
Race equivalence across distances and a pace converter between common units.
Training decision support
Performance translated into practical training intensity guidance, with an explanation of how the intensities fit together.
Why it belongs in an analytics portfolio
The interesting part is not the running. It is deciding what question each tool answers, how much methodology to expose, and where a calculation stops being useful without interpretation — the same judgement that decides what belongs on an executive dashboard.
It is also direct evidence of AI-assisted development: a working, published product designed and built end to end, rather than a prototype described in the abstract.
Scope
A personal product, not a client engagement. Guidance is derived from published running-performance methodology and is not medical or coaching advice.