Finances: money up to date without touching it

The problem

Almost everyone keeps their money in three places at once: their head, a folder of receipts and a spreadsheet updated whenever there is a spare moment. The bank statement says one thing, the spreadsheet another, and the feeling of being in control depends on the last weekend somebody spent reconciling them.

The important decisions get taken without the full picture. A public grant is missed because the window opened in one particular month and nobody wrote it down. A big purchase is made on impulse, or not made out of fear, because nobody really knows what the cash flow can take. And investing, where it exists at all, moves with the headlines.

What was built

First, the model of the data: every transaction, every rule and every deadline live in one place, ordered and dated. Then artificial intelligence working on top of that model, every day.

• Spending by API. Bank transactions arrive on their own, already categorised. Nobody transcribes receipts or types into a spreadsheet.
• An 18-month cash projection. It recalculates itself with every transaction. Before a decision, I can see what it can take.
• Grants and deadlines watched. Every grant has its status and its date; when a deadline gets close, the system flags it.
• Investing by written rules. The allocation is decided in advance and the system checks that it is being followed. Rules, not impulses.

Everything runs locally: the data never leaves my machine.

How it works day to day

In the morning, the previous day's transactions are already in and classified. The demo shows the scene with sample data: a supermarket under food, a petrol station under car, a chemist under health. Each one arrives on its own, with its category and its amount, without me touching anything.

When a decision involving money comes up, I look at the projection. The 18-month cash curve in the demo (staged, like everything in it) shows exactly that: where I am today and where I will be if I change nothing. I know what each decision can take before I take it.

Grants have their own living list. The demo shows three sample examples: a start-up grant already submitted with a favourable decision, an active contribution rebate and a tax relief whose window opens in October, with the reminder already set. That is the point: nothing depends on me remembering.

Even purchases still to come have a system: a wish list with priority and price, and when something is bought, the expense is recorded there and then. And investing follows its written rule (in the demo, a sample split between an index fund, a cash buffer and a pension plan), with the monthly contribution automatic on the 1st.

Result

I built the first digital twin for myself, and I keep it up to date daily with AI since June 2026. Today the system holds more than 400 log entries and more than 300 documents searchable in plain language, all 100% local and private.

In finances: bank transactions arrive on their own through an API, cash is projected 18 months ahead, grants and their deadlines are watched and flagged, and the investment rules are written down and the system checks that they are being followed. An automatic audit verifies every day that nothing is lost.

The same system covers the rest of life: projects with their next step, tasks and calendar in sync, contacts up to date, and health with a training plan on automatic progression, body composition measured to a protocol, nutrition logged and rest from the wearable connected since July 2026.