Medical Device Prototyping

The short answer. A prototype exists to answer one question at the lowest cost. We build the whole ladder: digital prototypes and simulation, proof-of-concept rigs, functional prototypes, a golden sample that looks and behaves like the production device, and batches of up to 200 units for bench, preclinical and clinical work. We make prototypes look like production early, because a sharp edge found after tooling can mean a new mold.

Types of prototypes and the question each one answers

Prototype	Question it answers	What it looks like	Who uses it
Digital prototype	Which concept is worth building?	CAD models, simulations, finite element analysis	The engineering team
Proof of concept	Does the principle work in the body or in the real environment?	A bench rig or test stand, often ugly	Founders, early investors
Functional prototype	Do all functions work together?	Real electronics and firmware in a printed or machined housing	Engineers, first users
Human factors prototype	Can people use it where it will be used?	Realistic size, weight, controls and labels	Clinicians and patients
Golden sample	Is this exactly what we will produce?	Production-intent design that fully meets the functional requirements	Investors, partners, the manufacturer
Pilot or clinical build	Does it perform in a study and on a production line?	A small batch built under the quality system	Study sites, the contract manufacturer

How we build and test prototypes

What prototypes teach that drawings do not

Test in the room the device is built for

The reason turned out to be simple: the lighting in an operating room is extremely bright, and under it the screen was unreadable. All the testing had been done under normal laboratory light. The device worked perfectly everywhere except the one room it was built for.

Watch real users

And we had a project where surgeons love the technical specs of the device, but the handle was wrong for how they naturally, you know, grip the instruments during like 4-hour procedure. So that’s something you can only learn by watching someone use it.

A prototype is not a production device

I would say one of the biggest gap at least from our perspective from engineering’s perspective is to understand that manufacturing is not just scaling the prototype.

That is why we make later prototypes look exactly like the device that will come off the production line, and why we involve the contract manufacturer before design freeze. Differences found after the pilot run, such as a sharp edge that needs a new radius, can mean redesigning the mold, one of the most expensive parts of production.

So what we are doing we are trying to engage our contract manufacturer early. So not after design freeze because a good manufacturer will look at your design and tell well I don’t know there are certain features that would add you don’t know 10 bucks per unit and 6 weeks to your timeline

How many hours prototypes take

Program	Prototype phase	Engineering hours
Rehabilitation chair electronics	First prototype: two boards and a remote	about 720
Rehabilitation chair electronics	Testing and iteration	about 500
Cold therapy device	Concept to five fully functional prototypes and a clinical trial	3,380 in 12 months
Glucose monitoring patch	Discovery to four prototype generations and clinical support	11,600 in 1.5 years

Prototypes and FDA rules

• Design controls. Since February 2, 2026 FDA design controls are ISO 13485 clause 7.3, applied through 21 CFR 820.10(c) to Class II and III devices, to Class I devices automated with software and to a few other listed Class I devices. Planning, verification, validation and transfer each need records in the design and development file, which replaced the old design history file.
• Prototypes in people. FDA’s QMSR preamble says prototypes used in clinical studies may be shipped under the investigational device exemption rules of 21 CFR part 812.
• Nonsignificant risk studies. A study of a device that is not significant risk counts as having an approved IDE once the IRB approves it and the abbreviated requirements are met. FDA will give a study risk determination through the Q-Sub program within 90 days.
• Significant risk studies. Implants, life-supporting devices and others with potential for serious risk need an IDE; the study may start 30 days after FDA receives it unless FDA objects.
• Early feasibility. FDA’s early feasibility study guidance allows first use in people before the design is final, generally in fewer than 10 initial subjects, with changes under a 5-day notice.
• Human factors. FDA finalized its guidance on human factors content in submissions on May 29, 2026, with three risk-based submission categories, and revised its human factors engineering guidance on August 3, 2026. Formative studies can use mock-ups and early prototypes; validation testing needs a user interface that represents the final design and at least 15 participants per user group.

Questions

Related pages

• Development for startupsWhat to build at each funding stage
• For manufacturersDesign for manufacturing and transfer
• All servicesElectronics, firmware, mechanics
• Case studiesEight projects in detail
• FDA QMSR in 2026Design controls under ISO 13485
• What development costs$125 per hour

Sources

Updated on October 9, 2026.