EmbeddedRelated.com
Blogs

The Backstreet Consultant

Ido GendelSeptember 18, 2023

In the uncharted land between Arduino-wielding kids and qualified electronics engineers emerged an entirely new market for embedded work. In this article, based on my personal experience and observations, I will attempt to outline this young market, the forces operating within it, and the kinds of people involved.

A short history

Up until fifteen years ago, give or take, embedded design was done exclusively by trained professionals: not just because of the required technical know-how, but also because of the expensive tools and processes of the trade. Very few individuals could afford them as part of a personal venture, and most of these were pros who knew exactly what they were doing to begin with. Then came Arduino (hold your eye-rolling, please – hear me out 😊), which sparked a huge positive-feedback loop of publicity, layman interest and cheap "Maker-friendly" products. Within a few years, the public perception of electronics shifted from an arcane high-tech industry into something that the kid next door could do. Also, more importantly for our topic, small-scale entrepreneurs started asking that proverbial kid to do electronics for them.

Many "traditional" engineers responded with contempt, then with some alarm. They felt that these new Makers were starting to steal their clients, and that they were giving engineering a bad reputation by creating half-baked, sub-standard systems which were worthy, at best, as proof-of-concept or as early prototypes. However, the former criticism missed the mark, and the latter was missing the point. I will illustrate this with a real-life example.

This article is available in PDF format for easy printing

Several years ago, right before COVID, Escape Rooms were all the rage in my country. Following the genre's initial success, it suddenly appeared as if every person with four walls to spare was bent on monetizing them in this manner. All it took – or so thought these low-budget dreamers – was a bit of themed decoration, and a bunch of riddles to solve. Such a riddle might utilize, for instance, a concealed load cell: put a specific weight on it, and the system will turn a little servo motor to reveal the next clue, plastic gem, whatever. Can a qualified engineer build such a device? Of course. Can a hobbyist Maker do it, for a fraction of the cost?

A new kind of embedded

Note, first, the clients in the above example: these are, on average, people with very little funds to spare. It would have been hard for them (financially or psychologically) to afford the time of a "proper" engineer, who's geared for developing mass-production and/or high-quality devices. Second, note their needs: the system does not have to be anything beyond a simple prototype. There's only one to be made, it doesn't need to adhere to medical, industrial or RF standards, and all it really takes is cheap off-the-shelf modules and a couple of lines of code (based on freely available libraries).

Other examples I've encountered include artists and art students, who need a one-off contraption for a short-lived art display; people who want a solution to a very specific, individual need in their home or workspace (this was huge before the days of commercial Smart Home devices); and beginner entrepreneurs who'd like to test their idea before they even present it to the first prospective investor. All share the same characteristics – a relatively low budget and, electronically speaking, modest requirements. These kinds of clients simply didn't exist before, certainly not in such numbers. They emerged and developed alongside the Arduino revolution, and their way of thinking is Arduino-oriented. They were never, and still aren't, the typical clients of professional embedded development firms.

To emphasize: In this line of work, the end goal is not a commercial product. Even when the client is dreaming of a gadget as ubiquitous as a smartphone, the current contract will end with a one-off, or an extremely small-scale production. Because of that, the cost of materials is negligible compared to the cost of labor (such as writing firmware). That is why Makers tend to use hardware which is a gross overkill for the tasks at hand, and rarely optimize for… anything, really. For a "good enough" device or two it is easier, faster, and cheaper to use a 35\$ Raspberry Pi with a few wasteful lines of Python code than to perfect a C/Assembly firmware for a 1\$ microcontroller on a custom PCB. The cynics might say that Makers use Raspberry Pis and Arduinos because that's all they know, but this is actually an aspect of the very same issue: even if they want to, no one is paying them to study new platforms, and in the short term – considering the lenient requirements of typical projects, as explained above - there will be no financial gain from such an investment.

That is not to say that all "backstreet consultants", as I call them, are limited to Arduinos. Far from it: they are a mixed bag of knowledge and abilities if there ever was one, ranging from kids who think that copy-pasting code examples can solve everything, to educated enthusiasts who will go into far deeper details than the average qualified engineer. The rates, too, vary widely, although they are still lower than those of the professional companies.

Takeaways

On the assumption that most readers here are qualified professionals, I'd like to address now the practical meaning of all this. To abuse the lingo, all these new projects are little more than spikes of noise for the pros; it's not the kind of work they're looking for. In fact, some engineers take it as a matter of professional pride not to work on the kind of quick-and-dirty devices that the clients want. The backstreet consultants serve, therefore, as the Low-pass filter, making sure that only the determined clients, with the proper plans and funding, will reach the next stage and seek the companies that can design and build an actual product. On the other hand, having passed through the purgatory of hobbyist contractors, these clients are now more knowledgeable, perhaps more suspicious, and they're used to lower rates; negotiating with them may become harder.

To bring it all together, in the last fifteen years a new embedded market has emerged, which is in fact new – not an invasion into or a leakage out of the old. This market is a playground for players with big dreams but small wallets. The classic standards of development - careful design, cost reduction, qualifications and so forth are mostly irrelevant, since the "products" are kept intentionally at the proof-of-concept or prototype level. This is a place where clients can get a quick, albeit imperfect, solutions for a reasonable price, or test the waters in preparation for the bigger game without risking too much in the process. Also, perhaps most important, it's a place where embedded hobbyists can gain experience and earn money to bootstrap their own embedded development business – a career path that was almost unimaginable two decades ago.


To post reply to a comment, click on the 'reply' button attached to each comment. To post a new comment (not a reply to a comment) check out the 'Write a Comment' tab at the top of the comments.

Please login (on the right) if you already have an account on this platform.

Otherwise, please use this form to register (free) an join one of the largest online community for Electrical/Embedded/DSP/FPGA/ML engineers: