Over a summer internship at a company I designed and built mechanisms for real machines — one went into production, another shipped production-ready.
A lathe-style can cutter: a brushless motor drives two silicone wheels through friction belts, spinning a can while a blade slices it open. To stop the thin aluminium wall deforming, I added a silicone-lined support roller underneath. V3 shipped with a BOM and laser-cut DXFs.
A universal test frame that holds AA, AAA and 18650 cells against spring-loaded pogo pins. I prototyped a lead-screw and a spring version; we chose the spring one for speed of use. It's in production.
The first machine I built — a can crusher — worked, but it solved the wrong problem. The client wanted a cutter. That mistake taught me to confirm the brief before designing, and it's why the cutter turned out right.