Product engineering · 15 years
Work directly with a senior product engineer with 15 years of experience. Choose a focused engineering sprint, validate an idea with a prototype, put AI to work inside your existing workflows, or reserve development time for your existing product.
Fixed prices. Scope agreed in writing before you pay. No retainer, no headcount, no long-term commitment.
A short form. What you want, what it has to do, and what success looks like.
I come back with what I would build and what it costs. In writing, before any money moves.
Deployed, handed over, and yours — with the source code and someone who can explain it.
Every package has a fixed price and a written boundary. Pick the one that matches where you are — and if none of them fit, say so in the form and I will tell you what does.
€199incl. Croatian VAT where applicable
A focused review of your idea or technical problem, ending in a written recommendation you can act on — with or without me.
€590incl. Croatian VAT where applicable
A full working day reserved for one clearly defined task on your product — a feature, a bug you cannot pin down, an integration, or a focused review.
€1,990incl. Croatian VAT where applicable
Take the one workflow your idea actually depends on and turn it into working, deployed software you can put in front of real people.
€790incl. Croatian VAT where applicable
I walk through your real processes with the people who run them, then send you a ranked shortlist of what is worth automating with AI, what is not, and roughly what each would cost.
€2,500incl. Croatian VAT where applicable
I look at how your team actually works, build AI into one or two workflows that genuinely benefit from it, and train the people doing the work so it keeps running after I leave.
Less structured, and usually less good value than a fixed package — but the right answer when the work genuinely does not have a shape yet.
from €400incl. Croatian VAT where applicable
A block of senior development time for an existing product, drawn down against scheduled work.
Not an exhaustive list — but if your project looks like one of these, we will get on well.
The spreadsheet works until three people need it at once. A small, well-shaped web application usually pays for itself within months.
Forms, availability, confirmations, reminders, payment. Bounded, well-understood, and the kind of thing that goes wrong quietly when it is built badly.
Enough of a working two-sided flow to find out whether anyone actually wants it, before committing to the full build.
Your customers want to see their own data. Your existing system was not built for that. This is mostly an integration problem, and integration problems reward experience.
Charging money correctly is deceptively hard: webhooks, retries, refunds, invoices, tax. I have shipped this several times, including the checkout this very page runs on.
Quoting, triaging support, pulling numbers out of documents, first drafts. The tooling is the easy half; building it into how the work actually happens, and teaching the people who do it, is the half that decides whether it sticks.
Slow, fragile, or written by someone who left. A focused day is often enough to find out whether it needs repair or replacement — and that answer is worth having before you decide.
Plenty of good options exist for building software — juniors, agencies and your own hires all do excellent work. This is for the situations where none of them is quite the right shape yet.
Good junior engineers are worth investing in, but that investment needs someone senior with time to guide it. If nobody on your side has that time right now, the work stalls and the mentoring doesn't happen either.
Here the decisions — data model, architecture, what to build first — come with the work rather than needing supervision.
Agencies are built for work large enough to support a team, and they're genuinely good at that. Below a certain size the coordination that makes them effective becomes overhead you're paying for without needing.
One person, one scope, no handover between roles. If your project outgrows that, an agency is the right next step and I'll say so.
Hiring is a significant annual commitment and a long process, and it's usually the right move eventually — but it's a decision made best once you know what you're actually building.
A fixed package costs what it says, ends when it ends, and leaves you with working software and a much clearer brief for whoever you hire next.
These are my own products, not client case studies — which means I can show you the whole thing rather than a sanitised summary, and I live with every decision in them.
A real-time multiplayer strategy card game with matchmaking, tournaments, achievements and a premium subscription. Runs continuously with real players and real payments — the kind of system where a bad state transition is visible to everyone in the room within a second.
A shared-world browser game with a persistent multiplayer canvas, upgrade progression and its own payment flow. Built to run cheaply and stay up without babysitting.
Single sign-on across every Proclive game, plus card checkout, Croatian CIS fiscalization and invoicing. Idempotent payment webhooks, server-verified amounts, audit trails — the unglamorous parts that decide whether taking money online actually works.
The page you're reading, the intake form, the approval workflow, the checkout and the invoicing behind it. Built in a few days, which is roughly what a Prototype Sprint gets you — so you can judge the pace and the quality from something you're already looking at.
I've also worked on plenty of things I can't put on a public page. Ask me in your project request and I'll tell you what's relevant to what you're building.
I am Igor. I have been building and shipping software for fifteen years — web applications, backends, games, payment systems and the unglamorous plumbing that holds them together. I run Proclive d.o.o. out of Duga Resa, Croatia.
Most of what I do is full-stack product work: taking something that exists as a description and turning it into a system that runs, handles real users, takes real money and can be maintained afterwards. I have done that as a solo developer, in small teams, and as the person other people escalated to.
The games at Proclive are my own products, which means I am also my own customer — I run the accounts, the payments, the invoicing and the fiscalization, and I feel every shortcut I take. That tends to make for conservative engineering choices, which is usually what you want from someone building something you will still own next year.
I use AI tooling heavily, and it is a real part of why I can deliver a working prototype in days rather than months. It is not the product. The product is judgement about what to build, an architecture that will not embarrass you in six months, and someone who takes responsibility for the result.
The whole process, including the parts most people leave vague.
Fill in the request form. It asks more than a contact form does, deliberately — the more precise you are about the one thing this has to do, the more useful my answer is. It takes about ten minutes.
Usually within two working days. I might come back with questions; that is a good sign, not a bad one. If I think a different package serves you better, or that you should not build this at all, I will say so.
For everything except the Clarity Session, I write down what I will build and what it costs, and you get to read it before anything is charged. If we change it afterwards, the old agreement stops applying and you accept the new one — you are never charged against a scope you did not see.
A secure payment link showing the exact scope and the exact price. Card payment, a proper Croatian invoice with fiscalization, VAT included in the advertised price. The link expires, and it stops working if anything changes.
We agree dates that work for both of us. Before the start date I will tell you exactly what access I need, because time spent waiting on a repository invitation is time neither of us gets back.
You get a status link for the engagement, and you get me on email. For anything longer than a day, you will see working software before the end, not just at it.
Deployed where you can use it, source code delivered, and a call to walk you through it. Then thirty days during which I fix anything that does not do what we agreed it would.
Me. There is no account manager, no bench of juniors, and nothing gets handed to someone else after you sign. You email me, I answer. That is the point of the whole arrangement.
Yes, as a tool, the same way I use a compiler and a debugger. It makes me faster at the parts that are mechanical. It does not decide your architecture, choose your trade-offs, or notice that the requirement you asked for conflicts with the one you asked for last week — that is what fifteen years is for, and that is what you are paying for.
Yes, and that is a package of its own — AI Workflow Implementation. It is not a strategy deck: I look at the work your team repeats, build AI into one or two workflows where it genuinely beats what you do now, and train your people to run and change it themselves. I sell no tools and take no commission, so if the honest answer is that a workflow should not be automated, you will get that answer.
That is normal and it is what the Technical Clarity Session exists for. An hour of hard questions and a written recommendation is a much cheaper way to find the shape of a project than discovering it halfway through building the wrong thing.
Yes, in full, once the invoice is paid. Pre-existing libraries and general-purpose components I bring with me stay mine, but you get an unrestricted licence to use them in your product. It is all written down in the service terms.
Fixed packages mean the exposure is bounded on both sides — a day is a day, a sprint is a sprint. Nobody is locked into anything longer than what was agreed, and I would rather tell you a project is not a fit than take the money and find out together.
Often, yes. A Senior Engineering Day is usually the right way to start: I spend a day in the code and tell you honestly what state it is in and what working on it would involve, before either of us commits to more.
Croatia, in the CET timezone. It matters for scheduling calls and not much else. Invoices are issued by a Croatian company with proper fiscalization — the registration details are in the footer.
Design, branding and copywriting, native mobile apps, ongoing maintenance, and production security or compliance audits, unless an accepted proposal says otherwise. Each package page spells out its own exclusions, and I would rather be blunt about them up front.
Ten minutes on the form, and you will have a real answer from someone who has built this kind of thing before — a scope, a price, and an honest opinion about whether it is worth doing.
Prefer email? igor@proclive.io
Not for you, but you know someone? Refer them and earn 10% — they get 10% off their first three payments.