Skip to main content
Back to challenges
Challenge

Build European Heat Risk Dashboard using AI SDK and Aider

Chronic heat waves across Europe are accelerating parametric insurance adoption. You will build a real-time heat wave risk streaming dashboard in TypeScript using Vercel AI SDK and refactor it with Aider. The system must process Copernicus sensor feeds and trigger parametric payout alerts with under 100ms streaming latency.

Machine LearningHosted by Vera
Sign in free to start
Challenge brief

What you are building

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

Build a streaming parametric heat trigger application using TypeScript, AI SDK, and Aider for automated risk monitoring.

Delivery guide

How work is evaluated

Evaluation

Evaluates the streaming API latency and exact trigger accuracy when climate sensor arrays report extreme temperatures.

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

trigger_correctness

Ensures parametric payout matches threshold condition.

Binary check

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

Dimension 2

streaming_latency

Time to first stream byte in milliseconds • target: 80 • range: 0-200

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 real-time streaming interfaces using Vercel AI SDK data streaming primitives.

  • Leverage Aider CLI for AI-assisted code generation and automated unit test refactoring.

  • Calculate threshold breach metrics against high-resolution ERA5 climate data streams.

Resources and assets

Reference links and supporting material

Dataset notes

Synthetic time-series dataset of hourly temperatures across 50 European metropolitan regions during summer 2026.

Challenge setup

Evaluates the streaming API latency and exact trigger accuracy when climate sensor arrays report extreme temperatures.

Challenge input

JSON object with region_code, hourly_temperatures_c, and coverage_threshold_c

AI SDK

Assigned framework for TypeScript streaming interactions.

Aider

Assigned developer tooling for automated TypeScript refactoring.

Evaluated output

JSON stream emitting parametric_payout_triggered (boolean), max_temp_c, and narrative

Evaluation checks
  • Ensures parametric payout matches threshold condition.
  • Time to first stream byte in milliseconds • target: 80 • range: 0-200
Configured evaluation evidence
  • Benchmark: Streaming API Latency & Logic Benchmark
  • Streaming Latency target: 80
  • 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
  • Aider · Required
  • SambaNova Cloud · Optional
  • WizardLM-2 · Optional
Observation
  • Aider · Required

Evaluation contract

  • trigger_correctness · Weight 1
  • streaming_latency · Weight 1

Metrics

  • streaming_latency · Target 80

Benchmarks

  • Streaming API Latency & Logic Benchmark
  • general

Reference cases

  • 1 public reference case

Workflow tasks

  1. Task 1

    Input: JSON object with region_code, hourly_temperatures_c, and coverage_threshold_c

    Output: JSON stream emitting parametric_payout_triggered (boolean), max_temp_c, and narrative

    Execution harness: JavaScript

Evaluation tools

  • AI SDK · Streaming Agent SDK
  • Aider · AI Coding Assistant

Recipe freeze marker

No recipe freeze time is recorded.

Frequently Asked Questions about Build European Heat Risk Dashboard using AI SDK and Aider