Banking software

// The problem

Business loan applications arrive by email, credit memos are written from scratch, and the borrower calls the branch to find out where their file stands.

// What we would build

A web portal for account managers and credit adjudicators with an iOS and Android app for business owners, carrying the file from application through adjudication and covenant monitoring, connected to the core banking system.

// What is in it

// Stack

• Next.js
• React Native
• PostgreSQL
• Core banking integration
• Credit bureau API
• AWS Canada

// The problem

Transaction monitoring produces a queue of alerts analysts cannot clear, and true suspicious activity waits behind thousands of false positives.

// What we would build

A custom model fine-tuned on the bank's own closed alerts and analyst notes, running on GPU servers in the bank's data centre so customer transactions never leave it, that ranks each alert and explains why for the analyst who decides.

// What is in it

// Stack

• PyTorch
• Python
• PostgreSQL
• Next.js
• Lenovo GPU servers on premises

// The problem

Several credit unions verify the same business members and their beneficial owners separately, and each keeps its own copy of documents that go out of date.

// What we would build

A private permissioned ledger where each participating credit union runs a node and records that a check was done, by whom and when, while the documents themselves stay with the institution that collected them. A consortium with one trusted operator could use a shared database instead.

// What is in it

// Stack

• Hyperledger Besu (private, permissioned)
• Next.js
• PostgreSQL
• AWS Canada

// The problem

Front-line staff search a shared drive of procedures, bulletins and rate sheets to answer a member's question, and often find an out-of-date version.

// What we would build

An assistant that answers from the current procedures only and cites the document and section, on an open-source model served in the bank's own AWS account in Canada so internal policy stays inside it.

// What is in it

// Stack

• Llama or Mistral (open-source)
• vLLM
• PostgreSQL with pgvector
• Next.js
• AWS Canada

// The problem

A credit union's board and management reports are assembled each month from core banking extracts, the lending system and spreadsheets, and two reports rarely show the same number.

// What we would build

A data warehouse in Canada that loads core banking, lending and card data every night into one model with agreed definitions, so dashboards and board reports all draw on the same figures, with access by role and a log of who read what.

// What is in it

// Stack

• PostgreSQL
• dbt
• Apache Superset
• Python
• AWS Canada

// The problem

A credit union's lending staff use an in-house desktop application for applications and approvals that only one contractor understands, and every change is slow and risky.

// What we would build

A modernisation in stages: we document what the old application does, write tests that pin its behaviour, then rebuild it screen by screen as a web app, running both side by side until the new one gives the same answers.

// What is in it

// Stack

• Next.js
• PostgreSQL
• TypeScript
• Playwright
• DigitalOcean