Continuous GTFS lets you transform, validate, and collaborate on your agency's GTFS data so you publish higher-quality feeds, more reliably, with less staff effort. Your scheduling vendor's raw export in – a clean, validated feed your riders can trust, out.
A set of scripts on your laptop, a scramble before every service change, and when something slips, your riders notice before you do. You've kept it running – you shouldn't have to keep doing it this way.
The same work, as a pipeline your team owns end to end.
A pile of one-off scripts, broken feeds and spec errors: stop_times.txt, hack_v2_FINAL.py, ⚠ 214 spec errors, feed_BROKEN.zip, manual_upload.sh, @#?!, fix_calendar.py, clean_stops.py.
One person's scramble, before every service change.
Bring the scripts off your laptop. Your feed-processing rules run automatically on every upstream change, in one place the whole team can work in.
Every run is checked against the canonical MobilityData spec, on input and output. Nothing broken reaches production – a staging gate stands in front of your live feed.
A review-and-sign-off step before you publish: see the validation results and a plain-language summary of what's in the feed – trips per day, per route – so more than one person can stand behind it.
The platform fetches from your upstream sources, runs your rules, validates against the canonical spec, and publishes through a staging gate to a CDN – continuously, on every upstream change. Every run is logged, diffable, and reversible.
Not a GTFS editor – you don't hand-author feeds or key in individual detours. It applies your automated rules to the data you already produce.
@step
def strip_rt_suffix(feed):
"""Trip IDs arrive with a vendor
suffix the realtime feed omits."""
feed.trips["trip_id"] = (
feed.trips["trip_id"]
.str.removesuffix("-RT")
)
return feedRunning in production at a major Pacific Northwest regional transit agency, validated against real agency data.
Built by the team behind some of the largest open transit data efforts in the country. Open-source core, so the pipeline is yours to keep, with or without us.
Walk us through how your GTFS gets from your scheduling system to your riders
Fifteen minutes.
Charlie Costanzo · Data Strategy Practice Lead
Email us directly at transit@jarv.us