Skip to main content
Back to challenges
Challenge

Cyber Cat Bond Trigger Monitor with AI SDK

Cyber insurance-linked securities rely on parametric loss triggers verified by real-time incident metrics and downtime telemetry. Build an interactive trigger validation microservice using Vercel AI SDK to stream real-time attack validation reports and trigger events. Achieve streaming responses with under 500ms time-to-first-token and zero schema errors.

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 fast streaming parametric trigger monitoring web API using Vercel AI SDK and TypeScript for cyber ILS instruments.

Delivery guide

How work is evaluated

Evaluation

Evaluates streaming latency (TTFT) and accuracy of cyber parametric payout triggers.

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

ttft_threshold_test

Time-to-first-token must be under 500ms.

Binary check

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

Dimension 2

schema_error_rate

Rate of JSON streaming parse errors • target: 0 • range: 0-0.05

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 agent responses using Vercel AI SDK `streamText` and `generateObject`

  • Parse real-time cloud downtime telemetry for parametric trigger evaluation

  • Achieve TTFT (Time-To-First-Token) < 500ms for live monitoring dashboards

  • Ensure 0 validation schema errors on parametric payout determinations

Resources and assets

Reference links and supporting material

Dataset notes

Telemetry stream logs from simulated cloud availability disruptions across major cloud providers.

Challenge setup

Evaluates streaming latency (TTFT) and accuracy of cyber parametric payout triggers.

Challenge input

JSON containing bond_id, affected_cloud_provider, global_outage_hours, and financial_loss_est_m

AI SDK

Vercel AI SDK provides fast streaming, object generation, and UI integrations in TypeScript.

Evaluated output

JSON with payout_triggered (boolean), payout_amount_m (float), and ttft_ms (float)

Evaluation checks
  • Time-to-first-token must be under 500ms.
  • Rate of JSON streaming parse errors • target: 0 • range: 0-0.05
Configured evaluation evidence
  • Benchmark: StreamingParametricBench
  • Schema Error Rate target: 0%
  • 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
  • VESSL AI · Required
  • Nvidia · Optional
  • Vercel AI SDK · Optional
Orchestration
  • Vercel AI SDK · Optional

Evaluation contract

  • ttft_threshold_test · Weight 1
  • schema_error_rate · Weight 1

Metrics

  • schema_error_rate · Target 0

Benchmarks

  • StreamingParametricBench
  • general

Reference cases

  • 1 public reference case

Workflow tasks

  1. Task 1

    Input: JSON containing bond_id, affected_cloud_provider, global_outage_hours, and financial_loss_est_m

    Output: JSON with payout_triggered (boolean), payout_amount_m (float), and ttft_ms (float)

    Execution harness: JavaScript

Evaluation tools

  • AI SDK · Agent Framework

Recipe freeze marker

No recipe freeze time is recorded.

Frequently Asked Questions about Cyber Cat Bond Trigger Monitor with AI SDK