Write-up in progress. The facts below are accurate; the problem, approach and results sections still need writing.
Personal finance applications
An expense tracker and a finance dashboard, built for my own use at university — tools I actually ran, not a tutorial exercise.
Link the GitHub repo, and a live demo if there is one. Buttons are
left out rather than pointed at # — a dead link reads
worse than no link. Markup to paste back in once the URLs exist:
<div class="project-links"><a class="btn btn-primary" href="URL">Live demo</a><a class="btn btn-secondary" href="URL">GitHub repo</a></div>
Problem
Two or three sentences. What were you actually trying to do, and why is it non-trivial? Name the hard part — if there wasn't one, this probably isn't a portfolio project.
Approach
Architecture, the key design decisions, and the constraints you were working under (compute, data, time, a marking rubric). Say what you chose and what you rejected — that's the part that shows judgement. One diagram beats three paragraphs.
Results
Numbers. Accuracy, latency, throughput, users, cost, before/after — whatever the project was actually optimising. Add a chart or a screenshot. If a result was mediocre, say so and say why; that reads better than silence.
What I'd do differently
Two honest sentences. What you'd change with hindsight, and what you learned the hard way. Almost nobody writes this section, and it's the one that reads as engineering maturity — don't skip it, and don't make it fake-humble.