We built a desktop AI app with a $0 ad budget. Here's what worked (and what didn't) in the first month

BrainWave Studio — Fri, 30 Oct 2026 07:08:37 +0000

We're BrainWave Studio, a tiny independent studio. This month we've been shipping Valaris — a Windows app (plus a web version for phones) that puts hundreds of AI models into one chat. You bring your own OpenRouter account, you see the price of a request before you send it, and your chats stay on your device.

No funding, no ad budget. Here's what we learned so far.

Why we built it

Paying for five AI subscriptions to use each model a few times a month felt wrong. Pay-per-request through one API is usually much cheaper — but the existing tools either hid the price, stored your chats on their servers, or were built for developers only. We wanted something a normal person could install and use.

What we shipped in the first month

Releases 1.5 → 1.6.3 — an update every few days, each one built and checked by our own pipeline (the app installs updates only after verifying our signature).
Agent mode — the model creates files, runs code, opens the page it built in a clean browser window, takes a screenshot and fixes what it sees. On a cheap model, 10 real projects cost $2.00 in total.
Valaris Hub — 30 ready-made skills (CV, business plan, SQL, homework help, trip planning). 3 free, all with Pro.
A site in 9 languages — 459 pages, each skill and each project with its own page.
~480 automated tests — boring, and the reason we can ship every few days without fear.

Decisions that saved us

Python + the system web view instead of Electron. The installer is ~17 MB, and all the logic lives in one language.
Licences without a licence server. Pro codes are signed with Ed25519 and checked offline. Nothing to host, nothing to go down.
No user data on our side. Chats live on the device; we never had to build (or secure) a database of other people's conversations.
Safety rules for the agent from day one. It only sees folders you pick, every write shows a diff and asks first, every change can be undone, and there is no delete at all.

What didn't work (yet)

Payments. Card payments for a small studio are hard; we started with Telegram Stars through our own bot. It works, but it's an extra step for buyers.
Unsigned installer. A code-signing certificate costs ~$120–170/year, so for now Windows shows a warning on first run. We know it costs us installs.
Building before telling. We spent weeks shipping before writing a single post. Our first article went out only this week — we should have started much earlier.

What we're trying now (all free)

Articles where the audience actually is: a business story on vc.ru, an agent/code story here on dev.to, a technical deep-dive on Habr.
Search: sitemap + Google Search Console, Yandex Webmaster, and IndexNow, so new pages show up in Bing and Yandex in minutes, not weeks.
Skill and project pages that answer real searches ("AI resume writer", "AI business plan") in 9 languages.
Product Hunt in early December.

Ask

If you've launched a desktop app without a signing certificate — did the Windows warning hurt you, and when did you buy one? And what was your first channel that actually brought users?

Links, if you want to look: valaris.app · the 10 projects the agent built: valaris.app/portfolio · every release with SHA-256 checksums: GitHub · our previous post: A cheap AI agent fixed 7 bugs for $0.01

— BrainWave Studio, the team behind Valaris

A cheap AI agent fixed 7 bugs for $0.01 — and did 9 more dev tasks. Total: $2.00

BrainWave Studio — Thu, 29 Oct 2026 17:56:55 +0000

We gave an AI agent ten tasks in plain English — a code review with fixes, a Telegram bot with tests, a data report from a CSV, a few web apps — and ran all of them on one inexpensive model. We didn't write or edit a line of code. Ten tasks took 52 minutes of model time and $2.00 in total.

Task 1: fix someone else's code

A folder with a small Python to-do app written by someone else, plus its tests. Most tests failed. The prompt:

Act as a careful senior developer — find all the bugs, including the ones the tests don't catch, fix them and write a report.

The agent ran pytest, read the code, found 7 bugs, fixed them without touching the tests and wrote a report:

The agent's report: 6 failed / 1 passed before, all 7 passed after — $0.01

One of the bugs is the classic that tests rarely catch:

- def __init__(self, tasks=[]):
-     self.tasks = tasks
+ def __init__(self, tasks=None):
+     self.tasks = list(tasks) if tasks is not None else []

Every TodoList() shared the same list — add a task to one, and it shows up in all of them. Another bug swapped day and month in a date format. Before: 6 failed, 1 passed. After: all 7 passed. One run, one minute, $0.01.

The setup

one model for everything — Claude Haiku 5.5, one of the least expensive in its family;
the agent can create files, run one program at a time (Python, Node, test runners), open a page in a clean browser window, take a screenshot and read the console errors;
we only looked at results and, when needed, asked for changes;
the price is the OpenRouter bill for all runs, fixes included.

The agent runs in Valaris, a desktop app (and web version) we build at BrainWave Studio: hundreds of models from OpenRouter in one chat, chats stored only on your machine.

All ten tasks

Task	Time	Cost	Runs
Fixing someone else's code	1 min	$0.01	1
Telegram bot with a shop inside Telegram	5 min	$0.10	2
Sales report from a spreadsheet	2 min	$0.07	1
Crypto dashboard for Bybit	4 min	$0.07	2
Online store	4 min	$0.07	1
Coffee shop website in 5 languages	7 min	$0.10	2
Habit tracker app	3 min	$0.05	1
Solar System quiz for kids	2 min	$0.03	2
Picture book “Pip and the Lost Star”	9 min	$0.19	4
Browser game “Gale the Hedgehog”	15 min	$1.31	5
Total	52 min	$2.00

"Time" is from sending the task to the finished result; "runs" is how many times we sent the task or asked for a change.

Task 2: a Telegram bot — with 34 tests nobody asked for

A flower shop as a Telegram Mini App: catalog, cart, promo code, checkout — $0.10

The bot plus a Mini App shop inside Telegram: catalog, cart, promo code, checkout, colors from the Telegram theme. Unprompted, the agent wrote 34 tests for the bot, and all of them pass. Five minutes, two runs, $0.10.

Task 3: a sales report from a CSV

848 orders → a Python script → a dashboard with charts and conclusions — $0.07

A CSV with 848 orders for a year. The agent wrote a Python script that computed revenue, average order value, top products and regions and the monthly trend, then turned the numbers into a dashboard with charts and plain-language conclusions. Two minutes, $0.07.

What we learned

Running code beats reading code. The cheap model got the review right because it could run the tests, not because it "understood" everything on the first read.
Screenshots catch what tests don't. For the web apps, opening the page in a clean browser window and looking at it fixed layout problems no test would flag.
Cost follows back-and-forth, not difficulty. One-shot tasks cost cents. The browser game was the most expensive ($1.31) because we built it in five small steps.
Sandboxing is the real feature. The agent only sees folders you picked; every file write asks first and shows a diff; any change can be undone; there is no delete; it runs one program at a time, never a shell, with a time limit and without your environment's keys and tokens.

The other seven

Crypto dashboard (read-only key via a local proxy, demo data) — $0.07

Online store: 16 products, filters, cart, 3-step checkout — $0.07

A coffee shop site in 5 languages — $0.10

Browser game — the most expensive, $1.31, built in 5 steps

What it doesn't replace

These are small, self-contained projects. Real codebases have more connected parts and need more review.
A human still decides when it's done.
All companies, people and data in the projects are made up — it's a demo, not client work.

Try it

All ten projects, live, with the exact prompts: valaris.app/portfolio
The app (Windows) and the web version: valaris.app
Every release with SHA-256 checksums and the changelog: github.com/ValarisAPP/valaris-releases

Valaris is free; agent mode is part of Valaris Pro, and models are paid at their OpenRouter price. What would you hand to a cheap model? Drop a task in the comments — we'll run the interesting ones and post the results.