Making anyone their own theme park tour guide.

NextRyd does not just rank rides by wait time. We contextualize each recommendation with guest location, walkability, park conditions, weather, schedule windows, and near-term wait forecasts to answer one practical question: what should you do next from where you are right now?

Geolocation-firstProximity and path-aware recommendation scoring

Near-Real-TimeNear instant wait time ingestion

Parquet datalakeHistorical training and analytics backbone

ML horizonsForecasts at 5, 30, and 60 minutes

What We Built

NextRyd is an end-to-end platform, not a concept demo. We run a production-style ingestion layer, persist structured facts and dimensions in a datalake, and serve recommendation logic through a modern .NET stack.

Unified ingestion workers for wait times, weather, schedules, holidays, and discovery updates.
Recommendation engine scores options using current location plus live and predicted conditions.
Context layer balances queue length with movement cost, timing constraints, and guest preference profiles.
Blazor experience is optimized for in-park, rapid decision cycles with continuously refreshed context.

Cloud Credits Request

We are seeking cloud support to increase throughput, reliability, and experimentation velocity as we scale our context-aware recommendation engine.

Compute: scale ingestion workers, feature pipelines, and model retraining for geospatially contextual scoring.

Storage: expand resilient datalake retention for multi-season and location-segmented model quality.

Platform Services: observability, managed APIs, and secure deployment automation for real-time operations.