Comment: Hi!
I am a new BSEE graduate who just started out as a post-silicon validation engineer for mixed-signal ICs. I saw your blog a while back and thought to myself, “Wow, I’d love to do what this person is doing, but I think they took a totally different track than I am on!” But a few days ago, I came across another post you made and saw that you also started out in post-silicon validation. What a change in perspective that post brought!
My question in short is this: what is your honest opinion about staying in post-silicon validation for the purposes of switching to one of the pre-silicon roles (design, verification, or even architect)? Is post-silicon validation a good starting place for someone who wants to grow their career into one of those roles, or would it be better to just go straight for a MS/PhD?
I saw your post about switching from GPU Validation to Architecture. It seemed like that individual was asking from the point of someone a few years down the line from where I’m at, so I’m wondering from the perspective of someone who has just gotten their BS and is still at an entry-level position.
It seems to me that this question is more common for those in our field than others, since others in my team have also expressed a similar interest. So I also ask this since perhaps this would be helpful to other people who are in a similar boat as I am.
Thanks so much for any help in advance! Appreciate your posts — they are very helpful to me being early on in my career.
Internet Stranger
Hi, Internet Stranger.
Some of the best architects and designers I’ve come across in my career started out in post-silicon. So to answer your question directly, post-silicon validation is both an excellent position in and of itself as well as a strong starting point for transitioning into a different role.
A word of caution is that it may be easy to get siloed into the role. What this means is that you may have limited visibility into the architecture and design or may just get exposure to the portions of the design that you are responsible for. The longer you spend in this mode the harder it gets to transition out of the position because your skill set and knowledge will become extremely specialized.
The remedy to this is to actively probe for knowledge and understanding outside of post-silicon validation. To be specific, this means reading the architectural specification as well as engaging with the architects and designers to understand the design holistically. The best post-silicon guys/gals are always pinging me and asking questions related to both validation as well as architectural design decisions. They call out bullshit and I love them for it because it makes for a better overall design. These are the types of post-silicon folks that we bring into the architecture or design side because they not only have a grasp of the architecture but they also have a structured way of poking holes in ideas. It just makes for better engineering.
As a personal anecdote, I was very much the post-silicon guy that engaged with the architects and probably bugged and pestered a fair amount. Trust me when I say that most good architects will welcome questioning because it means there is another pair of eyeballs scrutinizing a design that will eventually go to production. The first piece of architecture that I did was when I was a post-silicon engineer. I had built a relationship with one of the architects and established a rapport. I had volunteered to help out with designing and implementing a small part of one of our architectural features. A few months later I had officially transitioned into the architecture team.
Regarding the MS/PhD I will share three comments.
1. Don’t pay for it or at least try like hell to not pay for it. Seek out tuition reimbursement or try to find funding. Funding for an MS might be scarce but opportunities are out there.
2. Another degree isn’t absolutely necessary but I will say that it is rare to see architects with just a BS. In my team of fifteen architects we have just one that only did undergrad. Fun fact; started out in post-silicon.
3. I’d recommend assuming the post-silicon validation role and start building not just your validation skills but also your architectural chops. If you don’t see a clear path toward an architecture role then stick with it. If you sense resistance, consider graduate school.
Pado Pado 