Question from Anonymous Engineer 6
Your blog post is amazing. I recently left my job after 4 years in the VLSI/FPGA industry to pursue Masters in Computer Engineering. I wanted to learn more about Computer Architecture and the software aspects of it. One thing which was and has been bothering me is the thought that I am so late to this field. I am not good in software programming and OS/Compiler concepts. But after reading your post and your “generalist” terminology it kind of makes me feel that maybe I made the right decision, to pursue what I want to do irrespective of the stage in my life. My goal is to become a principal engineer someday and your blog post was really interesting.
I have one question but its not technical or related. I have suffered from depression in the past(during my undergraduate) because of which my grades tumbled. And as I am doing my Masters I feel that there is this constant pressure to perform well. It feels like if I spend a day not doing my research/studies then I start feeling guilty which keeps bothering my mind. I would like to ask how do you balance your desire to perform well in your job/profession along side taking breaks? I know you mentioned in your case that you did well compared to your peers, but what if I/anybody is not sharp or don’t get lucky in terms of opportunities?
Response from PadoPado
Hi. Your response really resonates with me as I had a really rough undergrad experience and it took until my second or third year to somewhat get my shit together. As a result I had a very sub-par undergraduate GPA and I’m sure you can relate how bad that feels and how hard it is to get that first job. My undergrad GPA was so poor that I had to take grad courses with non-matriculated status before even being accepted. My grad school experience was quite the opposite as I had matured a bit and was able to employ a more systematic approach to learning, studying and research and had good results. Also, in grad school I was trying to convince myself that I’m an “analog guy” (I’m not), and eventually pivoted away from analog design later in my career but am still conversational in analog topics. This lends itself to my generalist approach to my career and its worked out well because it’s not common for architecture, digital, firmware, or OS guys to know anything about transistors. I can kind of glue everyone together and work on solutions that work across these domains without being a de facto subject expert.
I’d be lying if I said that I wasn’t at the lab all the time during grad school. I guess you can say I front loaded a lot of effort into my grad school years and once I got my foot in the door with my first job offer I was able to finally relax. It was not an altogether pleasant time for me but looking back I don’t regret it. I’m not sure if it’s great advice but I’d say to tough it out because its not forever and it will open doors for you later. Or at least that was my experience.
Work is a lot less busy than school which is just strangely always artificially busy (hw, exams, research, grading, projects, thesis). Work has a good amount of downtime with product cycles and what not. Also, I have a bursty working style and am a firm believer of diminishing returns so I’m good about taking my breaks and letting my mind refresh.
Question from Anonymous Engineer 7
Hey, first thing thanks for doing this. I saw from your blog post that even though the majority of your expertise is in the digital domain you also seem to have done some work in analog/RF electronics related to PLL’s. I just wanted to get your opinion on the growth and future in the domain of analog IC especially RFIC design. Down the line do you see this to be still a thriving field or should one just opt to move to the digital domain now itself?
Response from PadoPado
Communication is going to rely on RF for the foreseeable future so I’m not concerned about it going away any time soon. RF guys have always been a rare breed among electrical and computer engineers and are compensated very well. The catch is that I think it’s hard to be good at RF. Some RF guys tell me its hard to do what I do but I disagree… lol. I think RF is way harder and those guys breath rarefied air. I’d say if you have the aptitude for it and enjoy it go for it. Just a side-note you probably need a PHD to get your foot in the door. It’s possible with a masters and a good project/thesis under your belt but PHD is the normal path.
Question from Anonymous Engineer 8
I have a question: Is computer architecture a dying field?
Response from PadoPado
Computer architecture is definitely not dying though newer things like deep learning and human computer interface have taken a lot of its shine. Google’s work with quantum computing is mind blowing and you can browse the computer architecture department websites at Wisconsin, UT, UIUC, GT to get a sense of what is cutting edge in more traditional computer architecture. As computing needs change architecture is going to have to change to accommodate the new applications so architecture isn’t going to die any time soon. Think of machine learning and GPUs/TPUs and how much progress has been made in this field recently.
Question from Anonymous Engineer 9
When applying for internships or full-time positions, what makes a qualified electric engineer in circuitry? As in, when I’m applying to EE positions involving circuity, what skills or experiences do I emphasis? And similarly for embedded systems positions, what do I emphasis? I also have a lot of research experience with the full project life cycle (proposal to dissemination). Should I highlight those qualities?
Response from PadoPado
If you’re talking about transistor level design at the entry level, I’d say a good test is if you can look at the schematic of the LM741 op-amp several weeks/months removed from your analog design class and just intuitively know how it works. This is easy work for analog circuits folks but will separate the “I kind of know circuits” folks from the folks that really get it. I’ve never interviewed for embedded systems positions so I don’t have anything in particular to add that isn’t already said on the Internet. With regards to projects/research, I love seeing it on resumes but make sure you know it inside and out as that’s ripe area for deep questioning during interviews.
Question from Anonymous Engineer 10
-what advice would you give to an electrical engineering student who is considering working in the semiconductor industry?
-where did you go to college and what did you study(EE,CPE,ECE)
-In your blog you stated that you mastered out of a PHD program, I’m assuming this meant that you dropped out once you met the requirements for a masters degree, why did you decide to do this?
-Is an MENG/MS/PHD worth the opportunity cost?
Response from PadoPado
Man, if you are a nerd and love computers/video games/tech working on CPUs and GPUs is a dream come true. There is just so much fun lab equipment and technology at your disposal and a bunch of smart and motivated people to work with.
I went to top 30ish schools for undergrad and grad. Outside of MIT, Stanford, Berkeley, and CMU which is the creme de la creme, the top 50ish schools are mostly a wash. They will all pump out some great talent.
I left my PHD because I didn’t get along with my professor and didn’t really jive well with the publish or die mentality because we were pumping out some bullshit just to publish. This was just my experience though and I saw firsthand other research groups that were run exceptionally well and the students were content. I’m still jealous I didn’t have that kind of experience.
If you are talking about just pure monetary opportunity cost I think a funded MS (rare) gives the best ROI. PHD tacks on an extra 3 to 4 years and think of the megabucks you could have been making in the industry. PHDs are for if you really love the academic environment and the research aspect of ECE. They don’t make much sense if you are judging purely from a financial perspective but they have a strong argument for life satisfaction if you have the right personality for it.
Pado Pado 