Our work

Two production AI platforms.

A small portfolio in number and a large one in substance. Both are multi-tenant, both are live, and both are in daily use.

Why two is enough

These platforms were built for financial advice — a heavily regulated industry with a cautious buyer and a compliance officer standing between the product and its users. The engineering problems they solve are not specific to that industry.

Sensitive data that cannot reach a model. Tenant isolation that has to hold at scale. Unstructured documents in, structured product out. Enterprise infrastructure wrapped so a small business can actually operate it. Those are the same problems in legal, insurance, healthcare, accounting and banking.

AdviceStudio: a generated advice presentation open in the editor, with a filmstrip of slides below.

Case study 01

AdviceStudio

Turns a dense 40-page compliance document into a client-ready interactive presentation in minutes — without the client’s personal data ever reaching the AI.

The redaction engine strips names, dates of birth, addresses, phone numbers, emails and account numbers before any AI call, then re-scans the result. Anything that slipped through is a hard stop, and the document is not sent.

Python 3.12 FastAPI PostgreSQL 17 ECS Fargate Stripe
Coremetryx: the analytics dashboard library, showing available dashboards with AI-enabled badges.

Case study 02

Coremetryx

Enterprise business intelligence, wrapped so a small business can use it — multi-tenant, branded dashboards built on Amazon QuickSight.

Namespace provisioning, per-tenant identity, role mapping and securely embedded dashboards — with a guarantee that no tenant can reach another’s data.

TypeScript React AWS CDK QuickSight Cognito

The client

Planfocus Consulting

Both platforms were commissioned by Planfocus Consulting, an Australian advice-technology firm serving financial advice practices and licensees. Stellar Corporate’s director, Stephan Mariani, founded Planfocus in 2015 and led its technology direction as CTO through the development of both products.

That combination — deep knowledge of the industry alongside the engineering to build for it — is why the products fit how the businesses actually work.

Building something similar?

Tell us the problem. We reply within one business day.