Skip to main content
Back to challenges
Challenge

Build Real-Time Catalog Sync Agents with AI SDK and Blaxel

Rapidly synchronizing inventory across decentralized ONDC seller nodes often results in race conditions and duplicate catalog updates. Build a serverless TypeScript routing workflow using Vercel AI SDK and Blaxel serverless agent runtimes to resolve catalog collisions instantly. Maintain less than 200ms end-to-end sync latency.

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.

Construct high-speed edge AI agents using TypeScript, Vercel AI SDK, and Blaxel to maintain real-time ONDC catalog synchronicity.

Delivery guide

How work is evaluated

Evaluation

Measures catalog conflict resolution latency and accuracy under high update throughput.

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

conflict_accuracy

Ensures latest timestamp record is preserved accurately

Binary check

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

Dimension 2

latency_ms

End-to-end execution latency in milliseconds • target: 200 • range: 0-500

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

  • Implement stream generation and function calling using Vercel AI SDK in TypeScript

  • Deploy agent functions to low-latency serverless runtimes using Blaxel

  • Handle incoming high-frequency ONDC protocol catalog mutations without loss

  • Optimize agent context windows for continuous streaming response evaluation

Resources and assets

Reference links and supporting material

Dataset notes

Synthetic time-stamped inventory mutations mimicking high-volume quick-commerce sales events.

Challenge setup

Measures catalog conflict resolution latency and accuracy under high update throughput.

Challenge input

JSON object with conflicting stock records

AI SDK

Provides structured tool outputs and streaming interfaces for JS runtime.

Blaxel

Executes low-latency serverless agents at edge nodes.

Evaluated output

JSON object with resolved master stock record

Evaluation checks
  • Ensures latest timestamp record is preserved accurately
  • End-to-end execution latency in milliseconds • target: 200 • range: 0-500
Configured evaluation evidence
  • Benchmark: Serverless Agent Sync Benchmark
  • Latency Ms target: 200
  • 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
  • Blaxel · Required
  • Vercel AI SDK · Optional
  • Zed · Optional
Orchestration
  • Vercel AI SDK · Optional

Evaluation contract

  • conflict_accuracy · Weight 1
  • latency_ms · Weight 1

Metrics

  • latency_ms · Target 200

Benchmarks

  • Serverless Agent Sync Benchmark
  • general

Reference cases

  • 1 public reference case

Workflow tasks

  1. Task 1

    Input: JSON object with conflicting stock records

    Output: JSON object with resolved master stock record

    Execution harness: JavaScript

Evaluation tools

  • AI SDK · TypeScript LLM SDK
  • Blaxel · Agent Hosting Platform

Recipe freeze marker

No recipe freeze time is recorded.

Frequently Asked Questions about Build Real-Time Catalog Sync Agents with AI SDK and Blaxel