Skip to main content
Back to challenges
Challenge

Build a Supply-Price Momentum Agent

In this challenge, you will architect a sophisticated real estate intelligence system using Mastra AI (TypeScript) to orchestrate data ingestion and automated machine learning workflows. You will focus on identifying 'divergent' housing markets where building permit growth (Census BPS) is failing to keep pace with home price appreciation (FHFA HPI). To optimize the forecasting component, you will integrate TPOT (Tree-based Pipeline Optimization Tool) via a Python bridge or microservice to automatically discover the best regression models for housing supply trends. The core of the challenge involves building a Mastra AI workflow that processes multi-geographic datasets, calculates momentum indicators, and utilizes an automated ML pipeline to predict future supply bottlenecks. You will define custom tools for the Mastra agent to query Census and FHFA APIs, then synthesize these signals into a 'Market Divergence Score' used by institutional investors to assess local market risk.

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.

Build an automated housing market analysis agent using Mastra AI for workflow orchestration and TPOT for hyper-parameter optimization of supply forecasts.

Delivery guide

How work is evaluated

Evaluation

The evaluation focuses on the agent's ability to correctly calculate momentum and the accuracy of the TPOT-generated supply forecast.

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

Mastra Workflow Integrity

Ensures the Mastra workflow executes all steps from ingestion to divergence reporting without failure.

Binary check

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

Dimension 2

Forecast Accuracy (MAPE)

Mean Absolute Percentage Error of the TPOT-optimized permit forecast. • target: 15 • range: 0-100

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 a multi-step data orchestration workflow using Mastra AI's Workflow engine

  • Automate the discovery of optimal forecasting pipelines using TPOT for time-series housing data

  • Develop a divergence detection algorithm comparing building permit velocity against price index momentum

  • Deploy a TypeScript-based agent that uses specific tools to fetch and transform Census BPS and FHFA HPI data

Resources and assets

Reference links and supporting material

Dataset notes

Sample data for 1 tasks

Challenge setup

The evaluation focuses on the agent's ability to correctly calculate momentum and the accuracy of the TPOT-generated supply forecast.

Challenge input

JSON containing historical BPS permits and HPI values for 24 months

TPOT

AI Workflow Automation · Workflow Runners

Mastra AI

TypeScript agent framework.

Bito AI

AI assistant for dev tasks

Evaluated output

JSON with divergence_score and trend_classification

Evaluation checks
  • Ensures the Mastra workflow executes all steps from ingestion to divergence reporting without failure.
  • Mean Absolute Percentage Error of the TPOT-optimized permit forecast. • target: 15 • range: 0-100
Configured evaluation evidence
  • Forecast Accuracy (MAPE) target: 15
  • 1 public reference case
Configured runtime
  • Docker execution harness

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
  • TPOT · Required
  • Mastra AI · Optional
  • Bito AI · Optional

Evaluation contract

  • Mastra Workflow Integrity · Weight 1
  • Forecast Accuracy (MAPE) · Weight 1

Metrics

  • Forecast Accuracy (MAPE) · Target 15

Reference cases

  • 1 public reference case

Workflow tasks

  1. Task 1

    Input: JSON containing historical BPS permits and HPI values for 24 months

    Output: JSON with divergence_score and trend_classification

    Execution harness: Docker

Recipe freeze marker

No recipe freeze time is recorded.

Frequently Asked Questions about Build a Supply-Price Momentum Agent