LlamaIndex & Galileo ONDC Merchant Inventory Risk Agent
Mitigate regional inventory stockout risks for quick-commerce catalog integrations across Open Network for Digital Commerce (ONDC). You will construct an agentic index that evaluates supplier health, demand signals, and supply disruption risks to trigger automated reorder workflows.
What you are building
The core problem, expected build, and operating context for this challenge.
Build a LlamaIndex agentic workflow using Galileo hallucination monitoring to analyze ONDC merchant inventory feeds and dynamically optimize fulfillment routing.
How work is evaluated
Evaluates the agent's ability to process catalog updates and generate accurate, un-hallucinated reorder actions.
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.
Schema Validation Test
Ensures output conforms to reorder action JSON schema
This dimension contributes its full weight only when the submission satisfies the requirement. Partial credit is not awarded.
Reorder Precision
Accuracy of trigger thresholds based on stockout math • target: 0.95 • range: 0.85-1
This dimension contributes its full weight only when the submission satisfies the requirement. Partial credit is not awarded.
What you should walk away with
Implement LlamaIndex agentic routing for dynamic multi-catalog ingestion
Integrate Galileo AI telemetry for monitoring hallucination rates in inventory recommendations
Construct a consent-aware supplier risk assessment pipeline for ONDC protocols
Automate reorder threshold triggers based on predicted regional demand spikes
Reference links and supporting material
Simulated ONDC grocery catalog feed with fluctuating stock levels and delivery lead times across multiple regional hubs.
Challenge setup
Evaluates the agent's ability to process catalog updates and generate accurate, un-hallucinated reorder actions.
Challenge input
JSON containing store inventory feeds and supplier lead times
LlamaIndex
Provides data indexing and agentic routing tools.
Galileo
Monitors hallucination and agent tool accuracy.
Evaluated output
JSON array of reorder actions with confidence and urgency flags
- Ensures output conforms to reorder action JSON schema
- Accuracy of trigger thresholds based on stockout math • target: 0.95 • range: 0.85-1
- Benchmark: SupplyChainAgent-Bench
- Reorder Precision target: 0.95
- 1 public reference case
- Python execution harness
- Python 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
- Llama Index · Optional
- Hamming · Optional
- Galileo · Required
- Galileo · Required
Evaluation contract
- Schema Validation Test · Weight 1
- Reorder Precision · Weight 1
Metrics
- Reorder Precision · Target 0.95
Benchmarks
- SupplyChainAgent-Bench
- general
Reference cases
- 1 public reference case
Workflow tasks
- Task 1
Input: JSON containing store inventory feeds and supplier lead times
Output: JSON array of reorder actions with confidence and urgency flags
Execution harness: Python
Evaluation tools
- LlamaIndex · Agent Framework
- Galileo · Evaluation & Observability
Recipe freeze marker
No recipe freeze time is recorded.