@C VanDorne I most certainly will follow up. Also, we are having an online chat on Friday (1:00 pm EST) and Harry Wintz of Ranstad will be joining (he is one of the recruiters interviewed for the article). I recommend you join the chat (text only) if you can as you can pose any questions direclty to him as well. Details here.
Please do, if not for your own peace fo mind, then for me and those like me. Although I'm employed, it seems that I fall into the "having trouble finding work" catagory that you describe, and I'd like to know where I stand.
@C VanDorne Thanks for your comments, I wrote the article and to be honest I had some challenges myself in parsing out what specific analog design skills lare in short supply. But let me take a stab at it: Designing IC's, specifically custom chips, that involve both analog and digital is the skill that is in short supply. And it's the board-level designers, who design using discrete components, who are in some cases having trouble finding work. i would argue that to get to the IC level, you need a really solid basic understanding of analog, but engineers I spoke with disputed that - saying that software tools were so advanced now one could skip that step. The other area (not discussed in the article) that is in hot demand is power. Sounds like i should be planning on a follow-up article!!!
I recommend staying in your current possition while you develop your own RF ideas/skills. Get a work area going in your garage or appartment and start experimenting with stuff. How about a security system in an unknown band, or an audio distribution system that includes music and room to room intercom, or anything that will keep you interested because in the end you'll get use it? (Just make sure your neighbor is cool with all the RF interfierence while you are refining your design.) Then the next time you walk into an interview you say "Yeah, okay, nobody has paid me to do this but check this out." Then lay out the schematics and start explaining to the interviewer how it works. You'll have gained experience and shown initiative. You'll get that job.
Herr Anderson, would you agree with me then that the schism between the two disciplines is that one primarily requires "discrete thinking" only, while the other requires "continuous thinking" in which the right answer could be the extremes but is more often somewhere in the shades of gray? And that finding that proper shade is often the trick? And that the people who can do this are rather special and should get paid tons more?
Okay, those last few were a bit gratuitous, but hey...a little fun now and again can't hurt.
" I think the shortages you mentioned were mostly in the IC design area..."
Elizabeth, I am glad to see that you mention this because I found that the constant use of the phrase "mixed signal" in this artical to be quite appropriate because that's about what I got from it. Is this void in analog expertise in the area of IC design, or designing with ICs? There is huge difference. If it's the former I'd say that is a niche area that very few of us explored after college, and we're talking about a pretty small pool to begin with. (At least in my experience, anyway. Btw, I looked into it and it seemed like a MSEE was about the starting point, so I passed.) But if the concern is about EE's who design with IC's then I'd say the pool is a little larger and would include your Power, Audio, Instrumenation and RF folks.
As for "digital designers" I have trouble with the phrase. It seems like a bit of an oxymoron to me because fo the way things are these days: Quick turn around and low-risk management require the demand for, and heavy dependance on, vendor supplied reference designs. When something doesn't work, lean on the supplier to fix their stuff. Proceed to software.
"When you work for big companies, you tend to get slotted. They don't move you around, and nobody cross trains," says Chenier.
We got a bottleneck on the career ladder and we keep pushing from the bottom when we should be pulling from the top. I see all these goofy STEM programs pushing basic engineering knowledge on children and the unemployed, when there is no shortage of basic level engineering skills. We have a shortage of expert level engineering skills, but we're not trying to get those skills to the engineering community in a manner that employers can use to vet job candidates. Once you graduate you are on your own. I've got a library full of books I've read on electronics subjects I've never really used in my career. I learned Verilog once. Used Icarus to work through the problems in the book. Wrote a seven segment decoder for some SLD tester I built and never used Verilog again. I can't blame employers for not taking something like that seriously if I put it on my resume.
What we need is some system of online logged verifiable education, where employers can look at the work you've done. A place where authors of textbooks can put accompaning course work that has a crowd sourced help forum. I know Coursera has an MIT basic electronics course. Do MIT graduated need a basic electonics course?
And we need to encourage employers to give employees time for education. Maybe payroll tax breaks for time spent on qualified education programs.
We need to increase engineering mobility, to move ideas around and increase innovation and expand the field. We all need take a step up the ladder, to make room at the bottom.
To surmise the article "Analog Engineers: Too Few or Too Many?" The POV of HR is snivel, whine, whimper way too few. The POV of analog engineers (and many engineers in general) stomaching the current job market, way too many.
Looking at the following quotes from the article:
"WIntz estimates that the typical time needed to fill an analog/mixed signal position is much longer today than it is for embedded software developers. "Within the first 24 hours that an embedded position is open, we easily have three to five candidates identified as a potential fit for the role. But with mixed signal design, it may be two to three days to identify even one qualified person," he explains."
"Part of the reason for the delay is the relatively small pool of qualified candidates and the fact that the skill set required has become much more niched, says Wintz."From what I see, I'd estimate only about one in 45 engineers might actually be qualified for one of these positions," he adds."
"Wintz admits that he's worked with a handful of clients that have had positions remain open for a year or more."
Using these quotes, one can guestimate using 1:45 to generate a "qualified" candidate and in 24 hours 5 "potential" (not "qualified") candidates are found, therefor about 225 candidates may have applied for that one embedded position within 24 hours. Now if this embedded position is open for say a month and the 225 applicants per day rate remains consistent (I wager the rate increases as news of the position spreads and those few "in the loop" have applied first), there could around 6,750 applicants of which 150 are potentially "qualified". Assuming even 10 interviews are granted for the Crème de la crème, means the odds of being hired could be 1:675 for that embedded position in a month.
Assuming the monthly rate is a third for analog engineers (2,250), the odds of being hired may be around 1:225 just for a month' opening. Running just the raw numbers for a whole year means up to 27,000 (600 "qualified") analog engineers may have applied for that one analog engineer position. That's a lot of engineers and might even be approaching the total supply of analog engineers. "the skill set required has become much more niched" and "they are asking for so many skills that basically no one on the planet would qualify" so is it any wonder no one can ever fill any position (threading the needle with a rope). Funny thing is, engineering schools hear there is a shortage of engineers from these lamenting employers and in response spew out hoards of new inexperienced degreed engineers to attack the trenches (WWI) only for the new engineers to find out that "qualified" (1:45) engineers is what employers really want and 44 new unqualified engineers are likely floating in the ether. Hey, don't worry, be happy. I bet those "qualified" engineers get several job offers a day and are doing great.
@ Navelpluis I find it pretty easy just to say that an engineer simply has to update his skills.
But update in which direction? One direction is for the discrete analog engineer to learn how to place familiar analog or logic onto silicon, for which the company usually has many others who can already do that and do not want to "waste time". There are other opportunies for skill updates that do not involve learning how to "place everything-on-silicon" .
For example, I once designed a digital biphase encoder/decoder with PLL clock recovery and violation counter to indicate link performance. I wanted to learn how to put this into an FPGA, but a junior engineer was assigned to this task because I was too busy with the rest of the analog design and the boss wanted it done fast, not have any time wasted by his one-and-only analog engineer learning how to program an FPGA. The faster logic functions were done external to the FPGA in discrete logic. The junior's FPGA worked first time, no bugs - but he could not comprehend how the circuit worked even after I tried to explain it to him. And I still had not acquired any knowledge of how to work with FPGAs.
Updating in a different direction involves things like PCB layout. Not ordinary layout, but how to squeeze multiple channels through the 1 mm distances between connector pins on a high-speed backplane. Two channels between each row of pins - without crosstalking. I dreamed up the physical structure that accomplished this, no help from computer simulations, the first prototypes were almost bang-on with their impedance, and the measured crosstalk was far better than hoped for. The bulk of the crosstalk was due to the card edge connectors. Interesting side-note was that several months later we ran the numbers through a beta-test crosstalk software package we were evaluating. The simulated tests showed that my scheme would never work. The vendor finally admitted that we had uncovered a bug in their software. But this kind of "skill update" does not carry much force on a resume.
Another form of skill update is learning how to push the boundaries in areas as yet unexplored, in both analog and digital. For this I refer you to
What are the engineering and design challenges in creating successful IoT devices? These devices are usually small, resource-constrained electronics designed to sense, collect, send, and/or interpret data. Some of the devices need to be smart enough to act upon data in real time, 24/7. Are the design challenges the same as with embedded systems, but with a little developer- and IT-skills added in? What do engineers need to know? Rick Merritt talks with two experts about the tools and best options for designing IoT devices in 2016. Specifically the guests will discuss sensors, security, and lessons from IoT deployments.