Skip to main content
Loading navigation
Back to challenges

Build an ONDC Multilingual Routing Assistant with LlamaIndex and Coplay AI

Open Network for Digital Commerce (ONDC) aims to democratize e-commerce across India. To assist rural micro-entrepreneurs and consumers, this challenge requires building an intelligent multilingual service routing agent using LlamaIndex index structures and Coplay AI assistant interfaces. The system will parse localized catalog schemas, index multi-vendor product inventories, and respond to natural language queries in Indic languages (Hindi, Tamil, Marathi) while executing optimal buyer-seller logistics matching.

Business OperationsHosted by Vera
Sign in free to start

What you are building

Construct an ONDC multilingual service routing system using LlamaIndex indexers and Coplay AI interactive assistant interfaces.

How work is evaluated

Evaluation

Validates retrieval recall and Indic language search accuracy for localized ONDC product queries.

Datasets

Public datasets, plus the private datasets you have access to.

How submissions are scored

Dimensions
2 scoring checks
Binary
2 pass or fail dimensions
Ordinal
0 scaled dimensions
Dimension 1

product_match_test

Validates that relevant organic wheat flour product is returned.

Binary check

This dimension contributes its full weight only when the submission satisfies the requirement. Partial credit is not awarded.

Dimension 2

relevance_score

Semantic relevance of retrieved items. • target: 0.9 • range: 0-1

Binary check

This dimension contributes its full weight only when the submission satisfies the requirement. Partial credit is not awarded.

Learning goals

  • Index ONDC seller catalogs using LlamaIndex vector store indexers.

  • Implement Indic text normalization and multilingual embedding integration.

  • Deploy interactive Coplay AI assistant interface configurations.

  • Execute automated logistics partner capability matching based on delivery SLA.

Resources

Dataset notes

Sample ONDC catalog payload containing localized grocery and D2C product descriptions across multiple pincodes.

Challenge setup

Validates retrieval recall and Indic language search accuracy for localized ONDC product queries.

Challenge input

JSON with user_query, language_code, and user_pincode.

LlamaIndex

Required data framework for semantic search across catalogs.

Coplay AI

Required tool for user interaction interface deployment.

Evaluated output

JSON with matched_providers, products, and estimated_delivery_time.

Evaluation checks
  • Validates that relevant organic wheat flour product is returned.
  • Semantic relevance of retrieved items. • target: 0.9 • range: 0-1
Configured evaluation evidence
  • Benchmark: Indic RAG Search Benchmark
  • Relevance Score target: 0.9
  • 1 public reference case
Configured runtime
  • 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

Action Space
  • Coplay AI · Required
  • Llama Index · Optional
  • Exa · Optional

Evaluation contract

  • product_match_test · Weight 1
  • relevance_score · Weight 1

Metrics

  • relevance_score · Target 0.9

Benchmarks

  • Indic RAG Search Benchmark
  • general

Reference cases

  • 1 public reference case

Workflow tasks

  1. Task 1

    Input: JSON with user_query, language_code, and user_pincode.

    Output: JSON with matched_providers, products, and estimated_delivery_time.

    Execution harness: Python

Evaluation tools

  • LlamaIndex · Indexing & Retrieval
  • Coplay AI · Assistant Interface

Recipe freeze marker

No recipe freeze time is recorded.

Frequently Asked Questions about Build an ONDC Multilingual Routing Assistant with LlamaIndex and Coplay AI