Skip to main content
Back to challenges
Challenge

AI SDK & Bolt.new Citizen LifeSG Workflow Automation

Build a TypeScript web application using Vercel AI SDK and Bolt.new to automate citizen life-event claim submissions and document parsing for Singapore LifeSG services. Achieve under 2 seconds latency per claim verification with 98% field extraction accuracy.

Workflow AutomationHosted 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 an interactive full-stack AI tool with AI SDK and Bolt.new for streaming LifeSG document processing.

Delivery guide

How work is evaluated

Evaluation

Evaluates parsing accuracy and latency of the AI SDK claim processing API endpoint.

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

Latency Threshold Test

Ensures processing completes within 2000ms

Binary check

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

Dimension 2

extraction_accuracy

Accuracy of extracted identity fields • target: 0.98 • 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

  • Implement streaming tools and structured outputs with Vercel AI SDK in TypeScript

  • Accelerate web UI development using Bolt.new AI engineering tooling

  • Design document parsing pipelines for Singapore public sector forms (LifeSG)

  • Optimize frontend component rendering speed for real-time verification status

Resources and assets

Reference links and supporting material

Dataset notes

Synthetic dataset of Singapore LifeSG form records and mock Singpass identity payloads.

Challenge setup

Evaluates parsing accuracy and latency of the AI SDK claim processing API endpoint.

Challenge input

JSON containing base64 string or plain text of a child birth certificate / grant application

AI SDK

Vercel TypeScript AI library for structured output and streaming

Bolt.new

Rapid full-stack web UI prototyping environment

Evaluated output

JSON matching expected citizen attributes with execution_time_ms

Evaluation checks
  • Ensures processing completes within 2000ms
  • Accuracy of extracted identity fields • target: 0.98 • range: 0-1
Configured evaluation evidence
  • Benchmark: WebArena / FormBench
  • Extraction Accuracy target: 0.98
  • 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

Environment
  • Bolt.new · Required
Action Space
  • Bolt.new · Required
  • Vercel AI SDK · Optional
  • Aim · Optional
Orchestration
  • Vercel AI SDK · Optional

Evaluation contract

  • Latency Threshold Test · Weight 1
  • extraction_accuracy · Weight 1

Metrics

  • extraction_accuracy · Target 0.98

Benchmarks

  • WebArena / FormBench
  • general

Reference cases

  • 1 public reference case

Workflow tasks

  1. Task 1

    Input: JSON containing base64 string or plain text of a child birth certificate / grant application

    Output: JSON matching expected citizen attributes with execution_time_ms

    Execution harness: JavaScript

Evaluation tools

  • AI SDK · Framework
  • Bolt.new · AI Engineering Tooling

Recipe freeze marker

No recipe freeze time is recorded.

Frequently Asked Questions about AI SDK & Bolt.new Citizen LifeSG Workflow Automation