01
Products
Small, focused software built under the CYBIND umbrella. Every product solves a problem I live with directly, and every product uses modern AI as real engineering material — not as a pitch-deck flourish.
An independent software studio · Lisbon, Portugal
CYBIND is my senior software practice in Lisbon. Twenty years of shipped work — streaming platforms serving 3 million viewers, property-management platforms, multi-brand consumer systems — now with modern AI as a core engineering material rather than a feature. Today the sharpest edge is software where correctness is the product: accounting, banking and tax, where a wrong number is the whole problem.
Shipped at, ongoing for, or built by
Financial & property systems
Streaming & OTT
Multi-brand CPG
Fixed scope, fixed price. You’re buying a defined outcome, not a block of someone’s calendar.
No timesheets, no ticket queues, no BD layer. Scope agreed up front, price agreed up front — and the senior engineer who scoped it is the one who builds it. It’s what one person with modern AI can now take on.
— How CYBIND works
From engineers I’ve shipped with
We worked together on the Locast.org project. Andrew brings an outstanding mix of strategic thinking, technical skills, and hands-on problem-solving — and a consistent high standard of quality.
Five years working with Andrew. Deep understanding of the trade-offs in high-traffic applications, project infrastructure, and system design. His AWS expertise is admirable.
Deep technical expertise plus exceptional communication. Andrew brings thoughtful architecture, reliable delivery, and the ability to turn ambiguous requirements into pragmatic solutions.
One of the most effective and reliable people I've met in my career. Strong technical skills, and the ability to set priorities and keep difficult projects moving.
Direct client references available on request.
The practice
I make software in two registers. The first is my own — products I own, ship, and support for the long haul. The second is for others — a small consulting practice for teams that need a senior principal with real AI fluency.
01
Small, focused software built under the CYBIND umbrella. Every product solves a problem I live with directly, and every product uses modern AI as real engineering material — not as a pitch-deck flourish.
02
Selective engagements for teams that want to put modern AI into a shipping product, or collapse what used to need three engineers into one senior principal with the right tools. Three to six months, retainer-based.
Working with me
Two ways in. Most people start small — a week of judgment, in writing — and go deeper only if it earns it.
Start here
Architecture audit · 1 week
A written deliverable, not a slide deck: your current stack and its real constraints, buy-vs-build calls, where AI genuinely helps and where it doesn’t, and a prioritised roadmap. No implementation, no long commitment.
Go deeper
Monthly retainer
For when the audit turns into build. An agreed scope and deliverables per month, 3–6 months typical, renewable month to month. Or a fixed-scope sprint for a single defined deliverable.
Start
Two calls, then an SOW
First call scopes the problem. Second call confirms fit and pricing. Then a one-page SOW, an NDA if you need one, and a start date — usually within two weeks.
Commitment
Scope, not hours
You’re not buying a share of my week — you’re buying a defined deliverable at a fixed price, with a date. I take on what I can deliver properly, and say so when I can’t. Current availability on request.
Pricing
Audit from €4K
Architecture audit €4–6K fixed, one week. Fixed-scope sprints from €8K. Retainers €12–25K / month depending on scope and depth. Invoiced from a Portuguese legal entity, in EUR or USD.
Procurement & ops
What I’ve shipped
The current edge is software where correctness is the product — accounting, banking, tax, and the property-management systems that run on them. The rest of the portfolio spans streaming at national scale and multi-brand consumer systems. Every block below is live somewhere, or has been.
01
Software that turns bank activity into a number someone can act on — reliably, in plain language, at the right moment of the month. TAXCLARA is the working proof: a Portuguese tax and accounting product with a tax engine encoding IRS, IVA and Social Security rules, PSD2 open-banking sync, AI transaction categorization, statement and receipt import, and a native iOS app. Designed, built, and shipped by one engineer.
Next.js · Prisma · PostgreSQL · SwiftUI · PSD2 open banking · Claude API
02
The operational core of a property-management business: double-entry accounting, owner and tenant ledgers, multi-tenant billing, reconciliation, leasing and booking flows — plus an internal platform that puts infrastructure health and portfolio health on one screen. Built for On Q Property Management — four markets across Arizona and Texas, in business since 2010 — including a large-scale migration off PropertyWare onto a modern, in-house platform, and current work applying AI to work-order estimation. The kind of work most teams outsource and most outsourcers don’t finish.
Double-entry accounting · owner & tenant ledgers · multi-tenant billing · reconciliation · AI work-order estimation · infra + business observability
03
Broadcasters and publishers who want to stream their own content, on every screen their audience owns, without paying an enterprise vendor a tax for every viewer. One platform, one brand, one bill. Ran nine live channels for a regional NBC affiliate from 2021 to 2026 — originally built under Jaybird TV, later under my ownership.
Web · iOS · Android · Apple TV · Android TV · Fire TV · Roku · Samsung Tizen · LG webOS
See the streaming & OTT practice → Read the Locast engineering case study →
04
One codebase, many brand voices. Full portfolio of sites for Ken’s Foods — Ken’s, Sweet Baby Ray’s, Sticky Fingers, Kogi Sauce, and the foodservice variants — plus the initial build of Green Valley Naturals. Every brand keeps its identity; every fix ships to the family at once.
Monorepo architecture · shared design system · headless content · per-brand deploy pipelines
Approach
My degree was in intelligent decision-making systems — neural networks, optimization, adaptive control. For a long time that felt like an academic curiosity. It doesn’t anymore. Modern AI now sits inside every engagement as a genuine co-worker: Claude in the editor, LLMs in production, agents running background jobs that would have been manual labor a year ago.
The practical effect is worth naming clearly. The shape of what one engineer can own has changed. Design, build, ship, maintain, support — functions that used to need separate people now compress into one practitioner with the right tools. That’s not a promise; it’s a description of how I work every day, on AWS infrastructure (Lambda, ECS, Bedrock, SageMaker) with the same observability and cost accounting as anything else in production.
What it doesn’t mean: dashboards labelled “AI-powered” for a press release, or replacing human judgement with prompts. What it does mean: a team that used to need three engineers to ship something now needs one senior engineer with real AI fluency. That’s what CYBIND is built to be. Full background on about.
Recent writing
What AI actually does for a senior IC working alone.
Streaming platform engineering at 3M-viewer scale.
Multi-brand CPG on a single monorepo, one engineer.
What two decades of production work taught me.
Why the studio is a single engineer on purpose.
The engineering backbone behind Ken's, Locast, WFMZ+.
Frequently asked
Contact
One inbox, no ticket queues, no BD layer. Tell me what you’re working on and what’s blocking it. If two decades of engineering plus modern AI can help, I’ll say so. If not, I’ll say that too. Most new engagements start with the one-week architecture audit — a small, fixed first step for both of us.
— Andrew Kovalenko, CYBIND