How the four fit together
Which one do you actually need?
Most enquiries arrive describing a solution — "we need an app" — when the useful conversation is one level up. These four services are not four products; they are four entry points into the same team, and which one you start from depends on how much of the problem is already settled.
Start with product development when the idea is still an idea
If nobody has decided what the first version does, building anything is premature. Product development is the scoping work: who the user is, what the smallest useful thing you can put in front of them is, and what the technical shape of that has to be so version two does not mean starting over. It ends with a plan, a prototype you can click, and a fixed quote for the build. Skip it only if you already have that plan.
Start with web development when the product lives in a browser
Marketing sites that have to rank and convert, admin dashboards that internal teams live in all day, storefronts with real checkout, and the Firebase or Node.js backends underneath them. The dividing line is not complexity, it is where the work happens: if your user opens a tab rather than an icon, this is the one. Performance and search visibility are part of the build here rather than a phase afterwards — this site is generated by the same approach.
Flutter for one codebase across iOS and Android, native Kotlin when a project genuinely needs it — background services, deep OS integration, hardware. This is also where store publishing lives, which is the part most first-time founders underestimate: review policies, release tracks, and the update cadence that keeps an app from quietly rotting after launch. We ship the release pipeline, not just the binary.
Assistants, document and receipt extraction, OCR, on-device ML. AI is a layer on a product, not a product: it sits inside a web app or a mobile app you already have or are building. The engineering that matters is the unglamorous part — what happens when the model is wrong, what a request costs at ten thousand users a day, and what runs on-device so it does not cost anything at all. We will tell you when a feature does not need a model.
Or hand over the whole thing
The common case is a mix: a Flutter app, a web dashboard for the team behind it, and one AI feature in the middle. That is one engagement with one team and one price, not three contracts. There are no hand-offs between a design agency and a dev shop, because there is no second company — the two of us are the people who scope it, build it, and answer the phone afterwards. The case studies show what that looks like end to end.
Not sure which line yours falls on? Describe the problem rather than the solution and we will tell you — including when the answer is that you do not need us yet.