Side-projects

Most building in public optimises for visibility rather than honesty. The version I find more useful is quieter: making decisions legible, tracking outcomes honestly, and creating a record that is worth something even if no one reads it.

Good enough is not a low bar. It is a precise bar. The threshold moves based on what you are trying to learn next, and getting it right is one of the harder skills in building anything.

I rebuilt rcordr with AI at the centre of the workflow, not the edges. The leverage is real, but the edge is not about tooling. It is about clarity of thought, fast iteration, and learning from others.

Building rcordr privately is joyful and restorative, but staying comfortable forever means it never gets tested. The challenge is letting reality in without losing the reason I started building in the first place.

Side projects don't need to become something bigger. For me, they're a way to let excess thinking escape—small decisions that are mine alone, without needing to turn into anything else.