Type-Safe Asset Management Platform with a CI-Gated Pipeline

Overview

A software studio was building an asset management platform for the real-estate sector — a system for tracking properties, portfolios and the operational data attached to them. We joined the engineering team to build the product's web application and stand up the delivery infrastructure around it, working daily in a small, quality-focused team.

This engagement is a good example of what disciplined product engineering looks like when a team commits to it: end-to-end type safety, tested components, static analysis gates, and deployments that are boring in the best way.

The Challenge

• A data-heavy domain. Property portfolios carry deep, relational data — assets, valuations, leases, documents — that needed careful modelling and fast, filterable views.
• Quality as a hard requirement. The product handled data that clients make financial decisions on. The team wanted test coverage, code review discipline and static analysis from day one, not bolted on later.
• A new codebase with no infrastructure. CI/CD, cloud environments and deployment workflows all had to be established alongside the product itself.

What We Built

Type-safe application stack

A React / TypeScript application with tRPC providing end-to-end type safety between client and server — API contracts enforced by the compiler, not by convention. Prisma handled data modelling against PostgreSQL, with Ant Design as the component base for the data-dense UI the domain demands.

Testing and quality gates

• React Testing Library + Jest for component and behaviour tests
• Storybook as the living component catalogue
• SonarQube static analysis wired into the pipeline, with quality gates enforced on every merge
• Daily code reviews and small, frequent commits — DORA-metrics-friendly delivery habits

Cloud and delivery infrastructure

We set up the initial AWS footprint — S3 for asset storage, RDS PostgreSQL, Lambda for background work — and built the GitHub Actions CI/CD pipeline that took every merge through tests, static analysis and deployment. Weekly production deployments were part of the routine from early in the project.

• React
• TypeScript
• tRPC
• Prisma
• PostgreSQL
• AWS
• Jest + RTL
• Storybook
• SonarQube

Outcome

• Shipped the asset management web application from early build-out through weekly production releases.
• End-to-end type safety via tRPC eliminated an entire class of client/server contract bugs.
• Test coverage, Storybook documentation and SonarQube gates kept the codebase maintainable as it grew — new engineers could contribute safely within days.
• The GitHub-to-AWS pipeline we established carried the team's deployments long after the initial setup.