These projects were completed within course requirements and shared team responsibility. The case studies preserve the design brief, methods, calculations, prototypes, test evidence, and unresolved limitations without assigning individual credit beyond what the record supports.
Coursework is presented at the team level unless the available record verifies an individual contribution.
Each entry leads to a full case study with the artifacts and caveats appropriate to that work. New projects can join the same system without implying identical methods, maturity, or evidence.
Course · Structures
Python-aided iteration, manual verification, fabrication, and a physical failure that exposed a connection the analytical model did not predict.
Key evidenceCalculated prediction increased from 240 N to 869.5 N; physical testing failed at a splice.
Course · Mechanisms + vision
A system concept combining drum feeding, Raspberry Pi camera classification, and a mechanical length sorter—with subsystem results kept separate from system-level claims.
Key evidenceThe mechanical length sorter reached approximately 98% at the tested angles; the classifier missed its target.
Course · Human-centred design
A student commuting problem translated into a foldable front-basket concept through stakeholder analysis, divergent concepts, calculations, proxy testing, and convergence.
Key evidenceA four-position proxy test informed a qualified preferred concept, not a validated product claim.
Available reports, code, calculations, images, and test records stay close to the claims they support. Missing evidence is identified rather than inferred.
Define the problem, requirements, assumptions, team setting, and constraints shaping the work.
Show the calculations, concepts, prototypes, tools, and trade-offs behind the chosen direction.
Keep reports, code, images, measurements, and observations close to the claims they justify.
Separate predictions from observations and identify unresolved risk, missing tests, or incomplete integration.
The personal collection extends the same evidence-led approach into self-directed experiments, software, and technical systems.