AI Macroeconomic Labor Market Impact Pipeline using Mastra AI and Novita AI
Singapore's AI expansion is boosting GDP growth, but its distribution across workforce sectors varies significantly. Policy makers require granular, real-time analytics on high-productivity sector creation versus displacement risks across local labor pools. In this challenge, you will construct an automated labor market telemetry pipeline using Mastra AI (TypeScript framework) running inference via Novita AI serverless APIs. The workflow analyzes national job posting databases, productivity metrics, and skills transformation data to produce dynamic labor displacement risk indexes across Singaporean industry verticals.
What you are building
The core problem, expected build, and operating context for this challenge.
Build an AI economic impact workflow with Mastra AI and Novita AI inference endpoints to model labor force productivity gains and displacement risks.
How work is evaluated
Evaluates workflow execution, Novita AI integration, and SSIC sector displacement calculation.
Challenge datasets
Public datasets, plus the private datasets you have access to.
How submissions are scored
These dimensions define what the evaluator checks and which criteria separate a passable run from a strong one.
sade_score_validity
SADE score must strictly range between 0.0 and 1.0.
This dimension contributes its full weight only when the submission satisfies the requirement. Partial credit is not awarded.
workflow_execution_ms
Total execution time for multi-step Mastra workflow • target: 1200 • range: 100-5000
This dimension contributes its full weight only when the submission satisfies the requirement. Partial credit is not awarded.
What you should walk away with
Build a Mastra AI workflow with steps for data ingestion, LLM analysis, and score synthesis
Integrate Novita AI fast inference endpoints for scalable processing of sector job descriptions
Calculate Sector AI Displacement Exposure (SADE) scores based on task vulnerability frameworks
Export structured policy intelligence dashboards for government advisory teams
Reference links and supporting material
Singapore standard industrial classification (SSIC) workforce dataset with AI task exposure indices and job vacancy figures.
Challenge setup
Evaluates workflow execution, Novita AI integration, and SSIC sector displacement calculation.
Challenge input
JSON containing SSIC sector code, job listings, and task breakdown
Mastra AI
Required TypeScript framework for multi-step workflow orchestration.
Novita AI
Required low-latency inference provider for execution.
Evaluated output
JSON containing exposure score, augmentation index, and primary policy recommendations
- SADE score must strictly range between 0.0 and 1.0.
- Total execution time for multi-step Mastra workflow • target: 1200 • range: 100-5000
- Benchmark: EconBench
- Workflow Execution Ms target: 1200
- 1 public reference case
- JavaScript execution harness
- javascript sandbox (unavailable on Versalist)
Summary shows the first task, up to three tools, and up to three evidence items. Open the technical recipe to view all configured items.
View technical recipe
Configured tools
- Mastra AI · Required
- Novita AI · Optional
- Aim · Optional
- Mastra AI · Required
Evaluation contract
- sade_score_validity · Weight 1
- workflow_execution_ms · Weight 1
Metrics
- workflow_execution_ms · Target 1200
Benchmarks
- EconBench
- general
Reference cases
- 1 public reference case
Workflow tasks
- Task 1
Input: JSON containing SSIC sector code, job listings, and task breakdown
Output: JSON containing exposure score, augmentation index, and primary policy recommendations
Execution harness: JavaScript
Evaluation tools
- Mastra AI · Agent & Workflow Framework
- Novita AI · Inference Provider
Recipe freeze marker
No recipe freeze time is recorded.