Rain-time ETA accuracy is pulling against the speed of the live prediction path.
Riders need a believable pickup estimate while the serving path must remain fast enough for every request.
Who you’d be doing this for
“The app said four minutes, then I stood outside for twelve and missed my train.”
Filipa Economou · Commuter
She uses the rider app after work and decides whether to wait once a driver has accepted.
What is at stake
Rain-affected post-match ETA error has reached 3.8 minutes, and riders are canceling after drivers accept. You have to weigh a targeted accuracy correction against live serving latency for every request.
Why it isn’t already fixed
Every obvious fix costs something else. That’s the part you’d have to decide.
- ETA accuracy vs. serving latency
- rain-segment correction vs. broad model stability
- rider confidence vs. update simplicity
- automated diagnosis vs. verified evidence
Why Lyft
Real-time ETA predictions shape whether rideshare riders wait for a matched driver or cancel the trip.
Written with these in mind
Not your kind of problem? 66 more at Lyft, or browse every organization.
This is the setup. The work is inside.
Running it puts you in the room: the full situation and its constraints, stakeholders who push back in their own words, and the decisions that are yours to make. What you produce becomes a Day One Plan — work you can show someone instead of describing.