Custom electronic product development
Any idea, brought to reality. Feasibility through prototype to a product you can manufacture.
Most products that fail do not fail at layout. They fail because a decision made in the first fortnight — the wrong processor, an unrealistic power budget, a certification requirement nobody checked — turns into a wall six months later.
We take electronic products from an idea to something you can manufacture, and we structure the work so the expensive decisions happen while they are still cheap to change.
The five stages
1. Idea and feasibility
You describe what the product should do, who uses it, where it lives, what it must cost and when it has to ship. We turn that into requirements, a block diagram, and an honest assessment of what is hard.
The output is a written feasibility report: the architecture we recommend and why, power and data budgets, the regulatory pathway, a risk register, and a cost model at your target volume. Fixed price, and yours to keep whether or not we continue.
This stage regularly changes the shape of a product. A radio requirement that forces a certification path, a battery life target that rules out the obvious processor, a unit cost that only works at ten times the volume you planned — all cheaper to learn now.
2. Design
Schematic capture, component selection, stack-up definition with the fabricator, and layout. Review gates at agreed points so you see the design as it develops rather than at a reveal.
Everything is version-controlled, so every revision is diffable and attributable — which matters when a product lives for a decade and the person who made a decision has moved on.
3. Prototype
Fabrication and assembly through EU suppliers, then bring-up: rails, clocks, boot, then every interface proven against the schematic in order.
You get a working board and a validation report that says what passed, what was modified to make it pass, and what is still unproven. The unflattering parts are included, because those are the ones that decide the next spin.
4. Certification
CE marking, and whatever else your market requires. We design for it from the start — EMC layout practice, band planning for radios, creepage and clearance where voltage demands it, and component grading appropriate to the risk class.
Testing runs through an accredited house. Our job is to make sure the chamber session is a formality rather than a gamble, and to turn any failures into design changes quickly.
5. Production
DFM and DFA review against your assembler’s real process capability, panelization, test strategy, and the documentation a manufacturer needs to build the product without phoning you.
Then handover: native source files, documentation, and a BOM with second sources on everything likely to go short.
What we are good at, and what we are not
| We do this | We partner or decline |
|---|---|
| Hardware architecture, schematic, PCB layout | Injection-moulded enclosure design |
| RF integration, power electronics, sensing | Certification testing itself |
| Board bring-up, MCU-class firmware | AUTOSAR classic, FPGA RTL |
| DFM, BOM engineering, supply-chain screening | Contract manufacturing |
| Requirements, documentation, traceability | App and cloud platform development |
A design house that claims the entire stack is telling you something about its sales process rather than its engineering. Where we bring in a specialist we say so, and they work under our accountability rather than becoming your problem to manage.
How you buy it
Fixed-price concept phase. The feasibility work, priced before it starts. The most common way to begin, and the cheapest way to find out whether we are a good fit.
Project delivery. Full development against a task-level estimate in person-days, with review gates.
Staff augmentation. Your project, your tools, your process, our engineers for a defined period — when the constraint is capacity rather than capability.
Where we are
Cluj-Napoca, Romania. EU-based, EU jurisdiction for your IP, GDPR-clean, and no offshore handover gap. Central European time, so a question asked in the morning is answered the same morning.
Project communication runs in English, Romanian or German — for DACH clients, technical reviews in German and a short flight rather than a twelve-hour time difference.
Common questions
Straight answers
How much does it cost to develop an electronic product?
For a self-contained connected device, budget roughly €25,000–€80,000 of engineering to reach a validated prototype, and expect the range to widen either way with complexity. A simple sensor node on proven silicon sits at the bottom; anything with a certified radio, a safety requirement or a custom module sits well above. We quote each phase separately with a task-level breakdown in person-days, so you approve a number before work starts rather than discovering one at invoice time. The feasibility phase is deliberately cheap because its job is to tell you whether the rest is worth doing.
How long does it take?
A realistic first pass is four to nine months from agreed requirements to a validated prototype. Roughly: two to four weeks feasibility and architecture, six to twelve weeks design, three to five weeks fabrication and assembly, then two to six weeks bring-up and iteration. Certification adds two to four months and runs partly in parallel. Anyone quoting six weeks for a full product is quoting for the layout and hiding the rest.
We have an idea but no specification. Is that a problem?
No — it is the normal starting point, and it is what the feasibility phase exists for. You describe what the product should do and who uses it; we turn that into requirements, an architecture, a risk register and a cost model. Plenty of good products start as a sketch and a frustration.
Do we own the design at the end?
Yes, completely. Full IP transfer including native source files, not just Gerbers and PDFs. EU jurisdiction. You should never be locked to a design house by a file format, and if you ever want to take the design elsewhere everything needed to do that is already in your hands.
Can you handle manufacturing too?
We take the design to a production-ready package and work directly with your chosen fabricator and assembler, including the DFM iterations. We are an engineering house rather than a contract manufacturer, so we do not mark up your build — we make sure the build works and hand you a supply chain you own.
What if the idea turns out not to be viable?
Then the feasibility phase has done its job and you have saved the rest of the budget. We would rather tell you in week three than in month eight. That answer comes with the reasoning and the numbers behind it, so you can take it to your own stakeholders.
Evidence
Where this has been applied
Hero: the sensor node installed in a ceiling or wall position, in a real room with real light. Context sells this one — a bare PCB says nothing about presence detection.
mmWave presence detection for building automation
Radar instead of PIR, because PIR cannot see a person sitting still.
Hero: system architecture diagram as a clean vector drawing, not a photo. Sensors on the left, compute in the middle, bandwidth figures on the arrows. This case study's value is the thinking, so show the thinking.
Edge-compute architecture for an unmanned surface vessel
Three architectures compared on real bandwidth budgets, so the platform decision was made on evidence rather than TOPS figures.