Built Environment Autonomous Work Dispatch Orchestrator
The physical built environment requires continuous operational monitoring and multi-trade work order routing. Build an asynchronous agent event loop using Celery and Redis to continuously monitor sensor streams, classify fault conditions, and dispatch trade specialized vendors automatically. Maintain continuous uptime processing over 10,000 daily telemetry events.
What you are building
Construct an asynchronous worker architecture using Celery and Redis for autonomous smart building dispatch workflows.
How work is evaluated
Evaluates task queue execution latency and correct trade vendor dispatch routing.
Datasets
Public datasets, plus the private datasets you have access to.
How submissions are scored
Trade Routing Verification
Chiller failure must be routed to HVAC Mechanical trade
This dimension contributes its full weight only when the submission satisfies the requirement. Partial credit is not awarded.
Dispatch Task Latency
Average task processing time in milliseconds • target: 50 • range: 1-500
This dimension contributes its full weight only when the submission satisfies the requirement. Partial credit is not awarded.
Learning goals
Build asynchronous Celery task queues backed by Redis message brokers
Parse incoming building sensor telemetry events concurrently
Route critical fault tickets to appropriate maintenance vendors based on trade taxonomy
Implement automatic retry policies and dead-letter queues for unhandled faults
Resources
Telemetry event payloads and maintenance work order logs from real estate operations.
Challenge setup
Evaluates task queue execution latency and correct trade vendor dispatch routing.
Challenge input
JSON with fault_type, building_id
Celery
Enables asynchronous background worker processing for IoT dispatches
Evaluated output
JSON with assigned_trade, dispatch_status, task_latency_ms
- Chiller failure must be routed to HVAC Mechanical trade
- Average task processing time in milliseconds • target: 50 • range: 1-500
- Benchmark: TaskQueueBench
- Dispatch Task Latency target: 50
- 1 public reference case
- Python execution harness
- Python sandbox (unavailable on Versalist)
Shows the first task, up to three tools, and up to three evidence items.
View technical recipe
Configured tools
- Zed · Required
- Lindy · Optional
- Voiceflow · Optional
Evaluation contract
- Trade Routing Verification · Weight 1
- Dispatch Task Latency · Weight 1
Metrics
- Dispatch Task Latency · Target 50
Benchmarks
- TaskQueueBench
- general
Reference cases
- 1 public reference case
Workflow tasks
- Task 1
Input: JSON with fault_type, building_id
Output: JSON with assigned_trade, dispatch_status, task_latency_ms
Execution harness: Python
Evaluation tools
- Celery · Distributed Task Queue
Recipe freeze marker
No recipe freeze time is recorded.