The people who run the diagnostic are the people who write the code behind every system on this site.
We take operations out of makeshift apps, spreadsheets and paper, and put them into a system of their own that the client’s team uses every day. There are two partners: Henrique is responsible for engineering, from São Sebastião, covering São Paulo and its northern coast; Igor is responsible for international operations, from London. What H&I knows was learnt where the system runs: inside the client’s operation.
The initials are not decoration. They are two people, and each does one half of the work.
We already ran businesses together before H&I, e-commerce and product personalisation, among others. H&I began when it became clear that what our clients were missing most was not better selling: it was an operation that stopped depending on people doing by hand what a system does on its own.
Henrique, Brazil
Henrique is responsible for engineering: he writes the code, deploys the system and keeps it running. He works from São Sebastião and covers São Paulo and its northern coast, in person when the project calls for it. The person who runs the review is the person who builds: you do not speak to a salesperson and then get handed to someone else.
Igor, London
Igor is responsible for international operations, from London: prospecting, proposals and the commercial relationship with clients abroad. He holds Italian citizenship, which keeps the door open to the European Union as well as the United Kingdom. He is the one in your time zone.
A business should not stop because someone did not turn up. That is the problem we solve.
Most of what we find is not a lack of technology, it is technology in the wrong place. The owner bought a system that does not fit, so the team invented the patch: the side spreadsheet, the notebook, the messaging group. The patch works for as long as the person maintaining it is there. The day they go on holiday, fall ill or resign, the patch goes with them, and the owner discovers the operation lived in someone’s head, not in the system.
Mission
To build bespoke systems that take on the manual work and keep the operation’s knowledge where it cannot walk out of the door: in the software. If someone is missing on a given day, the business keeps selling, invoicing and delivering.
How we measure it
Not the number of screens delivered. It is how much repetitive work left the team’s routine, how many typing errors stopped happening, and how much the owner can now see without having to ask someone.
What we do not do
Automation here is not about cutting jobs. It is about no longer spending your team’s best hours on typing, double-checking and chasing status, and giving those hours back to what only a person can do: serve the customer well and solve the hard case.
Where this shows up
It is what we are asked for most today. The system answers and qualifies on its own, takes orders, alerts the team when something stalls, and talks to whatever the company already uses, without forcing anyone to open yet another program to get their work done.
Three paragraphs, no curriculum vitae.
H&I was formed inside real operations, not in a classroom: a three-branch shop, a restaurant floor at eight in the evening, a building site mid-works. None of them had a technology problem. They had work that technology was getting in the way of.
An ERP and an installable app for a three-branch paint retailer: 62 database tables, 199 endpoints, roughly 82,400 lines of TypeScript. A restaurant till, across 15 published releases. A car-parts pricing engine rebuilt from scratch. A site-management system for a construction firm. Beyond those: WhatsApp automation, AI-assisted content production and e-commerce. And, as a product of our own, a platform that hosts several clients on one server with Docker, Traefik and Stripe.
Off-the-shelf software forces the operation to fit the software. When it does not fit, the team invents the patch: the side spreadsheet, the notebook, the WhatsApp group. We prefer it the other way round. The team already knows how the work is done; the software is what should follow. So we write the part that belongs to the client from scratch, and reach for off-the-shelf pieces only where they solve the problem without charging freedom for it.
We work on both sides, and the two are not the same thing.
It is written here on the page, not saved for the first meeting.
Based in São Sebastião, on the northern coast of São Paulo state, serving clients across Brazil. Where we can visit, we visit: seeing the dining room full on a Saturday night changes what gets built on Monday morning. Where we cannot, the work is remote, and what gets delivered is the same.
Igor is responsible for international operations, from London: bringing in the work, the proposal and the commercial follow-up. Engineering stays in Brazil. The H and the I in H&I are the two partners. London is a new front: what H&I takes there is the same method applied here, with the system written in English and Portuguese from day one.
H&I has no office in London. If your project needs someone inside your office every week, we are not the right choice. We say it now because finding out halfway through is expensive for both sides.
Four principles. Each one with the practical consequence it has for you.
You get the whole repository.
Source code, documentation and whatever is running in production stay with the client. If you decide to change supplier tomorrow, you need no permission from us and no export process. The full ownership and exit rules are written out on the How we work page.
The whole checklist is on the technology page.
fail2ban went in after we counted 346 SSH brute-force attempts in seven days. That is the kind of decision that comes out of a figure read in the log rather than a generic recommendation. Every item we apply is set out by name and in detail on the technology page.
The internet going down cannot stop the sale.
The restaurant till runs on the local network: if the provider drops on a Saturday night, the floor keeps selling and the orders upload themselves afterwards. In the retail chain’s app, the queue of changes is held on the phone itself and flushes on reconnect. We design for the worst day, because that is the day that decides whether the team trusts the system.
We read the diff before anything ships.
We run AI agents split by role, planning, architecture, UX, QA, security, red team, documentation, each isolated in its own Git worktree, with one shared board between them. We published that orchestration as an open skill, and anyone can audit what it does. The architecture decision is still ours, and the security review happens before anything goes live.
Whatever can be shown without exposing a client is public.
github.com/Riicke/SkillAgentTeam
The /agent-team skill is open source: AI agents orchestrated by role, isolated by Git worktree, with a shared board. It is our working process published in full, for anyone who wants to audit it before hiring us, or use it in their own team.
The workbench view: a new screen, a release going out, the before and after of an operation that was still running on paper.
The professional profile and the longer notes: why a technical decision was taken, and what went wrong on the way to it.
A 30-minute diagnostic, free and with no obligation: you leave with a written summary of your problem, and it is yours even if we never work together. You tell us what is still done by hand and where the operation jams. The person who replies is the person who builds.
suporte@myhidigital.com
For anyone who would rather put the context in writing first.
+55 12 99626-4723
For anyone who would rather talk before writing.
