Cross-Border Fast Fashion Supply Chain Inventory Flow & Routing Engine
Fast-fashion ultra-logistics companies like Shein rely on ultra-responsive inventory routing between Asian manufacturing hubs, regional distribution hubs, and target consumer markets. Managing rapid inventory turnover requires dynamic re-routing when customs delays or shipping disruptions occur. In this challenge, you will implement an autonomous inventory routing agent using Mastra AI in TypeScript. The agent monitors regional warehouse stock levels, evaluates multi-modal shipping costs (Air Freight vs Regional Sea Express), and triggers automated inventory re-balancing orders across Asian hubs.
What you are building
The core problem, expected build, and operating context for this challenge.
Build an autonomous fast-fashion supply chain inventory routing engine using Mastra AI in TypeScript.
How work is evaluated
Evaluates stock rebalancing decision accuracy and transit cost optimization.
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.
stockout_prevention_rule
Must select Air Freight when Sea Freight lead time causes stockout.
This dimension contributes its full weight only when the submission satisfies the requirement. Partial credit is not awarded.
logistics_cost_efficiency
Percentage saving against unoptimized express freight baseline • target: 18 • range: 0-100
This dimension contributes its full weight only when the submission satisfies the requirement. Partial credit is not awarded.
What you should walk away with
Construct multi-step inventory workflows using Mastra AI (`Workflow`, `Step`) in TypeScript
Model safety stock reorder thresholds based on demand forecast velocity
Evaluate dynamic trade-offs between Air Freight speed and Sea Freight cost
Generate structured purchase orders and cross-docking dispatch instructions
Reference links and supporting material
Fast-fashion e-commerce SKU sales history, warehouse inventory metrics, and Asian regional freight rate tables.
Challenge setup
Evaluates stock rebalancing decision accuracy and transit cost optimization.
Challenge input
JSON with current SKU stock, daily sales velocity, and freight options (Air vs Express Sea)
Mastra AI
Required TypeScript framework for multi-step inventory workflow composition.
Evaluated output
JSON with selected transit mode, reorder quantity, and total estimated freight cost
- Must select Air Freight when Sea Freight lead time causes stockout.
- Percentage saving against unoptimized express freight baseline • target: 18 • range: 0-100
- Benchmark: LogiBench-Inventory
- Logistics Cost Efficiency target: 18
- 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
- Modal · Optional
- Mastra AI · Required
- Giskard · Optional
- Modal · Optional
- Mastra AI · Required
Evaluation contract
- stockout_prevention_rule · Weight 1
- logistics_cost_efficiency · Weight 1
Metrics
- logistics_cost_efficiency · Target 18
Benchmarks
- LogiBench-Inventory
- general
Reference cases
- 1 public reference case
Workflow tasks
- Task 1
Input: JSON with current SKU stock, daily sales velocity, and freight options (Air vs Express Sea)
Output: JSON with selected transit mode, reorder quantity, and total estimated freight cost
Execution harness: JavaScript
Evaluation tools
- Mastra AI · Agent & Workflow Framework
Recipe freeze marker
No recipe freeze time is recorded.