Jetlag Coach

I've spent 20 years in tech. I know how to ship. And yet Jetlag Coach was harder to get out the door than I expected, partly because of life and partly because the problem turned out to be much harder than it looked.

The why

I fly between Vancouver and France several times a year. Over time I figured out that jet lag is actually solvable, noI expected the hard part to be the UI. It turned out to be the engine.

Generating a jet lag adaptation schedule sounds simple until you write the optimizer. The optimal plan depends on the full shape of the trip: the sequence of flights, the duration of each layover, how long you're staying at the destination. Change any of those and the right answer can flip entirely. A 36-hour layover is not a small layover doubled. A 4-day trip is not a short version of a 10-day trip. The plan has to be calculated end to end for the specific itinerary, not pieced together from generic per-flight advice.

And the constraints fight each other. The clock-shifting math will happily tell you to wake up at 3am, but the plan also has to leave you enough sleep at night to function, get you up early enough on departure day to actually make your flight, and stay inside what a real human will follow. The optimizer has to satisfy all of that at once.

I ended up building a full mathematical optimization engine. It's the part I'm most proud of, and the hardest to compress into a marketing headline.t just manageable. The science is well established: light at specific times shifts your circadian clock, the direction of travel matters, caffeine and melatonin have precise windows where they help or hurt. I started applying it, and I now land in France and sleep a full night on the first night. Most people don't know that's even possible.

That gap between what the science makes achievable and what people actually experience seemed worth building for.

The technical surprise

I expected the hard part to be the UI. It turned out to be the engine.

Generating a jet lag adaptation schedule sounds simple until you write the optimizer. The constraints conflict. Light exposure windows fight with caffeine cutoff times. Melatonin timing shifts based on direction of travel. The plan has to be genuinely different for someone staying three days versus three weeks. The model is asymmetric: eastbound is harder than westbound, and the adaptation rates differ. Multi-leg trips introduce cascading edge cases that most existing tools simply ignore.

I ended up building a full mathematical optimization engine. It's the part I'm most proud of, and the hardest to compress into a marketing headline.

The life part

I built this in 6 months of early mornings, late nights, and the occasional stolen weekend, while working full time as a CTO, with a toddler at home and a pregnant wife. There were weeks I barely touched it. There were weeks I couldn't stop.

I'm not saying this for sympathy. Plenty of people build under harder conditions. But it's the honest context for how this came together.

Where it stands

Jetlag Coach launched this week on iOS, Android, and web. It's bilingual, and I'm targeting the francophone market first, for reasons I'm happy to go into. In-app payments are live and the first users are in.

I'd love feedback from anyone who travels long haul. And I'm happy to go deep on the optimization engine with anyone curious about the technical side.