Part 8 of 11

Technology stack

Proven, open components chosen for unreliable links, spatial data and explainable AI. Nothing exotic, nothing that locks the operator in.

LayerChoiceWhy
PositioningMulti-band GNSS receiver with QZSS L1/L5/L6, MADOCA-PPP engine, QZNMADecimetre accuracy across Asia-Oceania with free corrections and authenticated signals
Tracker firmwareC on Zephyr RTOSFine-grained power management, secure over-the-air updates, mature GNSS and cellular drivers
Device messagingMQTT 5 over TLS, EMQX brokerBuilt for intermittent links; acknowledged delivery pairs with store-and-forward
Backend servicesPython 3.12, FastAPISame language as the AI models; asynchronous for thousands of connections
Event streamRedis Streams, moving to Kafka at large fleet sizesOrdered, replayable events so any service can recover without data loss
DatabasePostgreSQL 16 with PostGIS and TimescaleDBSpatial queries for geofences, compressed time-series storage for fixes
AI and analyticsscikit-learn, LightGBM, pandasIsolation forest for anomalies, gradient boosting for return time; explainable and light, no GPUs needed
MapsMapLibre GL with OpenSeaMap nautical layer and custom zone overlaysNo per-view map fees; buoys and seamarks shown as on a chart
Staff appFlutter for Android and iOSOne codebase, works through patchy beach connectivity with local caching
Control room dashboardFlutter web, sharing components with the staff appConsistent screens and faster delivery
NotificationsFirebase Cloud Messaging, WhatsApp Business Platform, Malaysian SMS gatewayChannels staff and owners already use
InfrastructureContainers on Malaysia-region cloud, defined in TerraformData residency in Malaysia, reproducible environments
ObservabilityPrometheus, Grafana, SentryTracker health, platform health and errors on one screen
DeliveryGitHub Actions with automated tests and staged releasesChanges reach the dock safely, one step at a time

Design principles

Store first, process second

Every fix is written to durable storage before any rule runs. A crashed service replays from where it stopped.

Assume the link will drop

Trackers buffer, the app caches, alerts fall back from push to SMS. Nothing depends on perfect coverage.

Explain every alert

Each alert says why it fired, in words a dock attendant understands, with the track that triggered it.

Configure, don’t code

Zones, rules, thresholds, rota and retention are settings. A new beach is a configuration exercise.

No lock-in

Open standards throughout. The operator can export all of its data at any time in standard formats.

How it scales

A jet ski riding ten hours a day at 10-second reporting sends about 3,800 messages a day including dock heartbeats. That is light work for the platform; the same design handles one beach or a national operator.

Raw storage is shown before compression. TimescaleDB typically compresses position data several times over, and fixes older than 90 days are reduced to trip summaries.

FleetMessages a dayRaw storage a monthHosting
20 units76,0000.2 GBSingle small server
100 units380,0001.1 GBSingle server plus replica
500 units1.9 million5.7 GBSmall cluster
2,000 units7.6 million23 GBCluster with Kafka