Skip to main content
Back to challenges
Challenge

Build an AI SDK Open Network Food Delivery Dispute Settlement Agent

Food delivery systems operating on open networks encounter order mismatch and delivery delay disputes. Build an automated dispute resolution agent using Vercel AI SDK that evaluates order telemetry and applies automated refund rules with 95% settlement accuracy.

Business OperationsHosted by Vera
Sign in free to start
Challenge brief

What you are building

The core problem, expected build, and operating context for this challenge.

Develop an AI SDK dispute resolution agent in TypeScript that audits order delivery logs and computes fair refund settlements.

Delivery guide

How work is evaluated

Evaluation

Evaluates dispute resolution recommendation precision and refund score accuracy.

Datasets

Challenge datasets

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

Evaluation rubric

How submissions are scored

These dimensions define what the evaluator checks and which criteria separate a passable run from a strong one.

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

refund_validity_check

Confirms dispute resolution approval flag

Binary check

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

Dimension 2

settlement_accuracy

Accuracy of refund calculation and fault attribution • target: 0.95 • 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

What you should walk away with

  • Build TypeScript dispute settlement workflows using Vercel AI SDK

  • Parse order tracking time-stamps and restaurant prep logs

  • Compute refund percentages based on ONDC SLA violation guidelines

  • Generate automated customer response messages and payout triggers

Resources and assets

Reference links and supporting material

Dataset notes

Synthetic food delivery tracking logs and customer complaint transcripts.

Challenge setup

Evaluates dispute resolution recommendation precision and refund score accuracy.

Challenge input

JSON order complaint object with timestamps

AI SDK

TypeScript structured object generation framework

Evaluated output

JSON settlement object with refund_percent and liable_party

Evaluation checks
  • Confirms dispute resolution approval flag
  • Accuracy of refund calculation and fault attribution • target: 0.95 • range: 0-1
Configured evaluation evidence
  • Benchmark: E-Commerce Dispute Settlement Benchmark
  • Settlement Accuracy target: 0.95
  • 1 public reference case
Configured runtime
  • 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

Action Space
  • Vercel AI SDK · Required
  • Coval · Optional
  • Ellipsis · Optional
Orchestration
  • Vercel AI SDK · Required

Evaluation contract

  • refund_validity_check · Weight 1
  • settlement_accuracy · Weight 1

Metrics

  • settlement_accuracy · Target 0.95

Benchmarks

  • E-Commerce Dispute Settlement Benchmark
  • general

Reference cases

  • 1 public reference case

Workflow tasks

  1. Task 1

    Input: JSON order complaint object with timestamps

    Output: JSON settlement object with refund_percent and liable_party

    Execution harness: JavaScript

Evaluation tools

  • AI SDK · Agent Framework

Recipe freeze marker

No recipe freeze time is recorded.

Frequently Asked Questions about Build an AI SDK Open Network Food Delivery Dispute Settlement Agent