Skip to content
Skip to content
CHIMBUK.

Try flutter, ai, pricing or sudoku

Money AI - Expense tracker app icon
Finance / ProductivityOwn product — live on Google Play, verifiable by anyone

Money AI - Expense tracker

Chimbuk · Contains ads · Everyone

  • 100+

    installs

  • Jul 6, 2026

    updated

  • Free

    core forever

Expense tracking with XP, levels and streaks — privacy-first: local data, optional user-owned Drive sync.

Screenshots — as published on Google Play

Open listing ↗
  • 100+ installs
  • Updated Jul 6, 2026
  • Everyone
  • Contains ads
  • No data shared with third parties
  • No data collected
Money AI store screenshot — The dashboard: this month at a glance — spent, left, streak. Published on Google Play.
Money AI store screenshot — Quick add — income or expense logged in seconds. Published on Google Play.
Money AI store screenshot — Custom categories with icons and colours. Published on Google Play.
Money AI store screenshot — Weekly and monthly spending insights, charted. Published on Google Play.
Money AI store screenshot — Budgets per month or category, with over-budget warnings. Published on Google Play.
Money AI store screenshot — XP, levels and achievement badges for real budgeting behaviour. Published on Google Play.
Money AI store screenshot — PDF expense report output for records or an accountant. Published on Google Play.
Money AI store screenshot — Backup and restore through the user's own Google Drive. Published on Google Play.
Money AI store screenshot — Dark mode and multi-currency — free, not paid tiers. Published on Google Play.
01 — The dashboard: this month at a glance — spent, left, streak

The actual pixels from the public Play listing (Google's CDN) — auto-advancing, pause on hover. Every number above is verifiable there; assets mirror into this repo at deploy.

The build

A budgeting app that treats retention mechanics as an honesty exercise: XP comes from real financial behaviours — logging, staying under budget — not from notification spam. No account required; backups go to the user's own Google Drive, which is how we design all data layers.

What we handled

  • Gamified XP/levels tied to actual budgeting behaviours
  • Local-first storage; user-owned Google Drive backup
  • PDF expense reports + CSV export for accountants
  • Multi-currency, recurring transactions, dark mode — free

The second Chimbuk product on Google Play: a budgeting app that uses game mechanics as positive reinforcement — XP only for real financial behaviour — while keeping every taka, dollar and riyal on the user's own device and their own Google Drive. Here is the full build story: the ledger engine, the privacy architecture, the export pipeline, and what each of those decisions should tell you about hiring us for a fintech-shaped product.

Why the second product is a finance app

A game tests performance; a money app tests trust. Rounding errors, lost transactions, a sync that overwrites the wrong month — in this category, bugs are not inconveniences, they are broken promises. Building one for ourselves was the deliberate choice to hold our engineering to the standard we claim when selling to banks-adjacent clients: correct numbers, on-device, recoverable, no excuses.

It is also the product where our positioning is easiest to verify. The Play listing declares no collection and no sharing, and the architecture makes that declaration structurally true — there is no server of ours to breach. Anyone can install it today and check what it does with their data.

XP that only rewards doing the thing itself

The gamification layer is an honesty exercise: you earn XP for logging transactions, staying under budget and completing daily challenges, and login streaks raise a multiplier capped at 2x. There is no mechanic for opening the app and nothing more; no streak-terror notifications at 11 PM. Badmarks unlock for budgeting, saving and consistency — achievements map to financial behaviour, not engagement theatre.

For client products this is the reusable lesson: retention loops work when they pay users in genuine progress. Every habit-shaped app we design gets this test — if removing a notification would drop usage to zero, the loop was spam, not product.

Local-first, user-owned cloud

Transactions, categories, budgets and streak state live in an on-device database. Backup and multi-device sync ride on the user's own Google Drive: the app writes an encrypted export into their storage, restores from it on a new phone, and our infrastructure never holds a copy. No accounts, no email capture, no third-party analytics reading financial context.

The engineering consequence is the selling point: zero storage cost, zero privacy-exposure surface on our side, and an offline mode that is total — the app cannot break because a server is down, because there is no server. For banks, NGOs and government-adjacent clients who ask 'where does our data go?', the answer is a design pattern we ship in production, not a slide.

Money math, done like an adult

Amounts are integer minor units end to end — no floating-point rupiah, no half-paisa drift; currency is per-account metadata, so multi-currency families and freelancers invoicing in dollars while spending in taka see coherent totals. Recurring transactions are generated deterministically against the ledger, so an edit to one occurrence never corrupts the series.

Export is where a budgeting app earns professional trust: the built-in PDF report generator produces clean, accountant-legible statements, and CSV export/import round-trips through Excel and Google Sheets without column soup. These formats are covered by golden-file tests in CI — if a release would mangle an export, it fails before it ships.

About the “AI” in the name

We are deliberate here because so many competitors are not: Money AI's intelligence today is rules-based — pattern-learned auto-categorisation and budget forecasting that run entirely on-device, in real time. Natural-language logging (“spent 350 at Kabilpara Grocery yesterday”) via a Gemini pass is queued for the next version and will stay optional, because a privacy-first ledger must not need the cloud to work.

That is the standard we apply to every AI feature we sell: the model earns its latency and its trust cost, or deterministic code does the job instead. Names can carry ambition; ship pages must carry accuracy.

The store-grade free experience

Dark mode, multi-currency, custom recurring rules — the paid unlocks in competitors' catalogues are baseline features here, free under a non-blocking banner ad. The Flutter codebase ships one UI language across phone and tablet with Material 3 theming, and release engineering follows the same pipeline as Sudoku Master: signed builds, staged rollouts, crash triage and a maintained “updated on” date, currently Jul 6, 2026.

The result is a product that does what its listing says — “track expenses, set budgets & save money — fun XP rewards for good habits” — and an engineering record, public on the Play Console-facing side, that any prospective client can audit.

Verified record

Platform
Android
Package
com.chimbuk.expense
Availability
Google Play (free, ad-supported)
Last update
Jul 2026

Update histories, store listing and privacy labels are public on Google Play — that's the point: every claim on this page can be checked without asking us for a screenshot.

Under the hood — what each box took to build

Craft, per feature

  • Log in seconds

    One-thumb quick add with remembered category, amount keypad and default account; entry-to-saved under two taps because friction is why budgeting apps die.

  • Categories that are yours

    Custom categories with icon and colour tokens stored locally; the chart layer re-renders instantly because aggregation runs on-device against the ledger.

  • Budgets with warning, not shame

    Monthly or per-category budgets with a live “left to spend” figure and an over-budget warning that fires before the breach, not after.

  • PDF & CSV that accountants accept

    A designed PDF statement generator and CSV round-trip for Excel/Sheets — both covered by golden-file tests in the release pipeline.

  • Drive sync you own

    Encrypted backup files in the user's own Drive — cross-device restore without our servers holding a single row of financial data.

  • Streaks & XP loop

    Deterministic XP rules with a 2x cap and achievement triggers tied to budgeting actions — engagement without dark patterns.

Commissioning this shape of product

The plan, as we would propose it

  1. 01 · Ledger design1–2 weeks

    Data model in minor units, currency rules, category taxonomy, what backup format survives the next three versions.

  2. 02 · Core app3–4 weeks

    Quick add, budgets, insights, recurrence engine — tested for totals correctness before anything is pretty.

  3. 03 · Loop & polish2–3 weeks

    XP/achievements, dark mode, tablet layout, PDF/CSV export, ad placement that never covers content.

  4. 04 · Trust & release1–2 weeks

    Drive sync encryption review, data-safety declarations, staged rollout, restore-from-backup drills on real devices.

What a client actually wins

  • No-server economics

    Local-first plus user-owned Drive means the product's cloud bill is zero and its breach surface is a filesystem the client already trusts.

  • Compliance as a feature sheet

    “No data collected” is an architecture decision — for regulated clients it turns months of paperwork into one diagram of the data flow.

  • Habit loops clients can reuse

    The XP system is a pattern library entry: streaks, capped multipliers and achievement triggers any retention-driven product can licence into its roadmap.

  • Professional outputs, free

    PDF statements and clean CSVs make a consumer app double as a small-business record — the same generator code reskins into client-branded reports.

What is not true (yet)

  • Early audience again: 100+ installs, no review wall to quote — the proof here is architecture, not popularity.
  • No bank aggregation by choice: connecting accounts is a trust and compliance programme we have not built, and we would rather say so than half-build it.
  • Natural-language logging via Gemini is on the roadmap, not in the store build — the page says “queued”, not “shipping”.
  • Android-only today; iOS is a Flutter flip plus a store-process project, not a rewrite.

Testimonial slot — reserved, not invented

The listing has no review wall to quote yet — so there is no quote here. Real, credited words move into this card as players leave them.

Sponsored placement — reserved, disclosure first

Placements & partners

This is where a genuinely used tool or partner link would live — hosting, device labs, our travel products. It ships empty and labelled on purpose: when a sponsored link arrives, it arrives wearing this disclosure, never dressed up as editorial.

Answers where they belong

Questions this project answers

Want Money AI-class engineering for your idea?

One brief in, the composer arrives with your platforms pre-ticked — Android, iOS, web, admin.

Where does my financial data live?

On your device. Backups go to your own Google Drive as encrypted files — Chimbuk's infrastructure never holds a copy, which is why the Play listing can honestly say nothing is collected.

Is it actually free?

Yes. Dark mode, multi-currency, recurring transactions and exports are all free; a non-blocking banner ad funds it. No feature sits behind a paywall in the current build.

Can I sync between phones?

Through Drive: sign in with your Google account inside the backup flow, restore on the other device. No Chimbuk account exists to forget the password of.

Can I hand the records to my accountant?

That is what the PDF report and CSV export exist for — they are tested every release to stay legible in Excel and Google Sheets.

Why is it called Money AI if there is no chatbot?

The intelligence today is on-device, rules-based categorisation and forecasting. A Gemini-powered natural-language entry mode is queued for the next version and stays optional — the ledger must work with zero cloud.

We want a fintech-shaped app. Is this the template?

Closest thing to a proof: local-first ledger design, user-owned sync, export pipelines and store-grade release engineering — scoped through a Discovery Sprint, delivered in 7–11 weeks for an MVP.

Begin the ascent

Everything on this page is checkable.

Store listing, code craft, and the way we scope — click through and verify, then send us the brief.