Case Study

Real Estate Investment Data Pipeline (Next.js + Python)

Full-Stack Analytics Dashboard for Brickbro — transforming raw property data into actionable insights using Airflow ETLs and Atomic Design.

ProductionFull-Stack + Data EngineeringArvin Kent Lazaga
ETL
Pipeline aggregating market, property, and geospatial data sources
Dual
Database persistence — MongoDB (documents) + PostgreSQL (relational)
Maps
Google Maps API integration for geospatial property context
Atomic
Design system applied to Next.js frontend reporting dashboard

Overview

What was built

Brickbro Property Analysis is a commercial real estate evaluation system built to help investors make informed decisions using structured data and analytical reporting. The platform aggregates property-level metrics, market indicators, and geospatial data to generate structured insights and practical investment recommendations.

As a full-stack developer working closely with engineers and designers, I contributed to backend services, data evaluation logic, and frontend reporting interfaces that transformed raw property data into meaningful decision-support outputs.

The system addresses a core pain point in commercial property investment: data is fragmented across multiple sources, manual assessment is time-consuming, and there is no consistent analytical reporting layer. The platform consolidates these inputs into a structured evaluation pipeline and surfaces results through a clean reporting interface.

System Architecture

Data Pipeline & Evaluation Flow

End-to-end flow from raw data ingestion through ETL processing, backend evaluation services, and structured report generation to the investor dashboard.

01

Data Sources

Property data feeds, market indicators, and geospatial APIs provide the raw inputs for the evaluation pipeline.

Market APIs · Property Data Feeds · Geospatial APIs

02

ETL / Data Processing Layer

Apache Airflow DAG orchestration drives Python-based extraction, transformation, validation, and dual-database loading of aggregated data.

Apache AirflowPythonDAG OrchestrationData Validation
03

NestJS Backend Services

Modular NestJS services expose RESTful endpoints, enforce business logic, and integrate with external data providers through clean adapter layers.

NestJSTypeScriptREST APIAdapter Pattern
04

Analysis Engine

Structured evaluation logic processes aggregated property and market data to produce consistent, comparable investment metrics and scoring.

Evaluation LogicInvestment MetricsMongoDBPostgreSQL
05

Report Generation Layer

Processed evaluation outputs are structured into analytical reports with actionable investment recommendations and geospatial context.

Structured Reports · Investment Insights · Google Maps API

06

Next.js Frontend Dashboard

Reporting dashboard built with Next.js and Atomic Design — presenting investment insights, property comparisons, and geospatial data to investors.

Next.js · Atomic Design · Serverless · Data Visualisation

Core Processing Layer
Integration / Interface Layer

Engineering

Technical highlights

Backend Data Aggregation Services

Built NestJS services to aggregate property-level metrics and market indicators from multiple external data sources through clean adapter layers — keeping integration logic isolated from evaluation business logic.

Structured Evaluation Logic

Implemented structured property evaluation logic that processes aggregated data and produces consistent, comparable investment metrics — transforming raw inputs into decision-support outputs.

ETL Pipeline Contributions

Worked on Apache Airflow DAG-based pipelines for property and market data ingestion. Python extraction and transformation tasks load processed records into MongoDB and PostgreSQL via Prisma.

Google Maps API Integration

Integrated Google Maps Places API (Autocomplete) and geospatial context into property records — enriching evaluation data with location intelligence for investment analysis.

Atomic Design Frontend Architecture

Built and structured the Next.js reporting dashboard using Atomic Design principles — atoms, molecules, organisms, templates — ensuring consistent, reusable UI components across all reporting views.

Serverless Component Adoption

Introduced serverless components for stateless compute tasks where they improved efficiency and deployment flexibility without adding operational overhead to the main backend.

Stack

Technologies used

NestJS · TypeScript · Next.js · Python · Apache Airflow · MongoDB · PostgreSQL · Prisma · Google Maps API · Serverless · Atomic Design · Docker

Need a similar system?

Let's scope your integration and automation requirements.

Discuss your project