Rescue, Project work, Full-time
I take products from idea to live on the App Store and Google Play — design, native code, backend, monetization and the store listing — without a team behind me. Alongside that, core-team experience building a complete custom Android launcher that other studios’ products, with millions of installs, are built on.
Anything can be built now. Apple took 557,000 new app submissions in 2025 and review went from a couple of days to a couple of weeks. Getting approved, staying approved, and still earning in eighteen months is the part that got harder — and that is the part I have ten years of.
What I can build for you
From an app stuck in review to one that does not exist yet — end to end, in your repository from day one.
Rejected in review and you cannot work out why. A half-built codebase someone walked away from. An app quietly losing money on a misconfigured ad setup. This is the work I am most certain about, because I have been on the wrong side of most of these myself.
Google Play since 2016 · App Store since 2021
Tell me what’s stuckNot a prototype and not a design file — a real app, in review, then in front of users. Native iOS and Android, the backend behind it, analytics, monetization, and the store listing that gets it approved. You describe the product; I hand you the link.
11 apps live · 17 store listings · publishing since 2016
Tell me about the ideaAgent harnesses, multi-agent orchestration, memory and retrieval, MCP servers so your tools are callable by an agent. Built the way I build my own: unattended, tested, and trusted with work that matters.
ImageForge — 13 local models, HTTP API + MCP server
Explore AI workHow I work
The same four steps keep eleven products maintained without the quality sliding — and they are how I’d work on yours.
Reading the code, the console, the store listing — the real source — before proposing anything.
The smallest change that proves the approach, in front of you. Rough on Tuesday beats polished and wrong on Friday.
In milestones you can see and use, each at a fixed fee agreed before it starts — through review and live on the stores.
Anything done twice becomes a script or an agent, and decisions are written down with the measurements behind them.
Background
Much of my experience runs deeper in the platform than app work usually goes: core-team work building a complete custom Android launcher — home screen, app drawer, widget hosting and the widget picker, drag-and-drop for apps and widgets — the layer every app sits on top of, for products with millions of installs. Alongside that, AOSP and Android TV set-top boxes, where you are working with the platform itself rather than the SDK.
On my own account I’ve published more than forty apps over the years. That catalog was reset when Play policy enforcement removed the older ones, and I rebuilt from zero — the eleven live today are the rebuild.
I mention it because it is where the store-policy depth came from. If you are hiring someone to get a product onto a store and keep it there, that experience is the point.
3.5M+combined installs
My experience there is on the core team that built the launcher itself — home screen, app drawer, widget hosting and the widget picker, and drag-and-drop for apps and widgets, everything a stock launcher does. Other studios used it as their base, customizing the −1 screen, the pull-down and widgets with what we gave them. Four of their products are still live, under four separate developer accounts — the install counts are right there on the listings, so you can check every part of this yourself.
SwipeTap Apps
Avalon Solutions LLC
App Lab Studios
Clarity App Crafting
Install counts read from the live Play listings on 2026-09-12
To be exact about it: I didn’t build these apps, and I didn’t build the launcher single-handedly either. My part is on the core team that built the launcher they run on.
What working with me is like
The hard part of software is not starting. Eleven products are in front of real users, still updated, still passing review. Most portfolios are prototypes; this one has ratings.
Including when it was my mistake and what it cost. Bad news early is cheap; bad news late is not, and you cannot make good calls on a filtered picture.
I check the thing itself rather than the status describing it. A 200 response can be a login wall; a rendered input can be disabled. If I say it works, I looked.
Design, native code, backend, AI tooling, store ops. Nothing waits on a specialist to become available, and nothing gets lost between two of them.
For recruiters and hiring managers
The track record above is what I’d bring to a team: someone who has owned products end to end, shipped them repeatedly to two stores, and built the AI tooling that made that pace possible. The roles that fit best are AI and agentic engineering, native mobile, and founding-engineer positions where one person carries a product.
Where I’m strongest
Quick answers
I don’t publish rates, because the honest number depends on what the thing actually is and I have never been able to guess that from a description. Tell me what it is and I’ll give you a fixed number.
Fixed fee per milestone, agreed before it starts. Not hourly. You know what the next piece costs before I begin it, and an overrun is my problem rather than your invoice.
You pay when a milestone is delivered and you have approved it. Not before. If what arrives is not what we agreed, it is not done yet.
Everything lives in your repository from day one. Your accounts, your keys, your store listings. Stop after any milestone and you keep every line of it and can hand it to anyone.
No retainer, no minimum, no notice period. I have no payroll to cover, so I do not need to lock you in to cover it.
I have no client logos to put here. The work has been my own products and core-team platform work, and neither comes with a client testimonial. So instead:
I take one build at a time; short reviews and rescues I can usually start sooner. I reply to every email within one business day, including the ones I’m turning down.
Get in touch
A couple of sentences: what it is, where it is now, and what is stopping it.
Ready when you are
A TestFlight link, a rejection notice, a repo, a screenshot of the console — whatever the real thing is.
I’ll come back with what I think is actually going on and whether I’m the right person for it. If I’m not, I’ll say so and tell you what to look for instead.
Reply within one business day. No pitch deck, no sequence.
