Explore
Pre-canned queries against the live database. The SQL is shown alongside the result.
All queries hit the public API at quantumlane.io/api.
Daily ingestion volume (last 14 days)
Record counts per feed per day, across both real-time feeds.
SELECT received_at::date AS day,
'ttc.vehicle_positions' AS feed_key,
COUNT(*) AS record_count
FROM realtime.vehicle_positions
WHERE received_at >= NOW() - INTERVAL '14 days'
GROUP BY received_at::date
UNION ALL
SELECT received_at::date AS day,
'ttc.trip_updates' AS feed_key,
COUNT(*) AS record_count
FROM realtime.trip_updates
WHERE received_at >= NOW() - INTERVAL '14 days'
GROUP BY received_at::date
ORDER BY day DESC, feed_key;
Loading...
Vehicles currently in service
Most recent position per vehicle, looking back at most 5 minutes.
SELECT DISTINCT ON (vehicle_id)
vehicle_id, route_id,
ST_Y(location::geometry) AS latitude,
ST_X(location::geometry) AS longitude,
speed_mps
FROM realtime.vehicle_positions
WHERE received_at > NOW() - INTERVAL '5 minutes'
AND vehicle_id IS NOT NULL
ORDER BY vehicle_id, received_at DESC
LIMIT 20;
Loading...
Vehicles on a single route (the 504 King)
Same query, filtered by route_id.
This is the human-name โ route_id bridge the upcoming MCP server is built around.
SELECT DISTINCT ON (vehicle_id)
vehicle_id, route_id,
ST_Y(location::geometry) AS latitude,
ST_X(location::geometry) AS longitude,
speed_mps
FROM realtime.vehicle_positions
WHERE received_at > NOW() - INTERVAL '5 minutes'
AND vehicle_id IS NOT NULL
AND route_id = '504'
ORDER BY vehicle_id, received_at DESC;
API: GET /v1/routes/504/vehicles
Loading...
Per-feed freshness
The latest snapshot per feed โ the same telemetry the
Freshness page renders.
Note the latest-per-key pattern (DISTINCT ON).
SELECT DISTINCT ON (feed_key)
feed_key, status, lag_seconds,
record_count_5min, record_count_1h, last_record_at
FROM ops.freshness_snapshot
ORDER BY feed_key, snapshot_at DESC;
Loading...
Recent ingestion runs
Per-run records-written telemetry across all assets.
SELECT run_id, asset_key, started_at, status, records_written
FROM ops.ingestion_runs
ORDER BY started_at DESC
LIMIT 20;
Endpoint and schema are wired, but per-run telemetry isn't being written yet โ
this populates once the ingestion-run recorder is in place. Shown here as a
planned observability surface rather than live data.