Airport People-Mover Anomaly Monitoring with Mastra AI
Contractors confirmed the Al-Maktoum airport automated people-mover (APM) award. Build a TypeScript telemetry agent using Mastra AI to ingest real-time sensor streams (vibration, door cycle count, motor temp), detect anomalies, and auto-dispatch maintenance tickets. Maintain an anomaly detection accuracy above 95% with zero false negative critical warnings.
What you are building
The core problem, expected build, and operating context for this challenge.
Build a Mastra AI TypeScript workflow with built-in memory and tools to monitor telemetry and manage predictive maintenance for airport transit systems.
How work is evaluated
Evaluates APM telemetry anomaly detection and ticket generation accuracy.
Shared data for this challenge
Review public datasets and any private uploads tied to your build.
How submissions are scored
These dimensions define what the evaluator checks and which criteria separate a passable run from a strong one.
critical_detection
Ensures temps over 90C trigger immediate high priority ticket
This dimension contributes its full weight only when the submission satisfies the requirement. Partial credit is not awarded.
detection_precision
Precision of detected anomaly categories against benchmark • target: 0.95 • range: 0-1
This dimension contributes its full weight only when the submission satisfies the requirement. Partial credit is not awarded.
What you should walk away with
Initialize Mastra AI agent with memory storage and dynamic tool definitions in TypeScript.
Stream APM train telemetry data through Mastra workflows.
Define heuristic and LLM-assisted anomaly evaluation tools.
Automate maintenance work-order creation in response to persistent telemetry alerts.
Reference links and supporting material
Telemetry logs simulated for airport automated people-mover system bogies and door actuators.
How this agent runs
Evaluates APM telemetry anomaly detection and ticket generation accuracy.
Challenge input
JSON telemetry object with car_id, motor_temp_c, vibration_hz, and door_cycles
Mastra AI
Offers robust TypeScript agentic and memory tooling.
Evaluated output
JSON containing status flag, detected anomalies, and generated ticket
- Ensures temps over 90C trigger immediate high priority ticket
- Precision of detected anomaly categories against benchmark • target: 0.95 • range: 0-1
- Benchmark: IoTTicketBench
- Detection Precision target: 0.95
- 1 public reference case
- JavaScript execution harness
- javascript sandbox (unavailable on Versalist)
View technical recipe
Configured tools
- Mastra AI · Required
- Metaflow · Optional
- Hamming · Optional
- Mastra AI · Required
Evaluation contract
- critical_detection · Weight 1
- detection_precision · Weight 1
Recipe state
This is a preview. The configuration can change before the evaluation recipe is locked.