Are you enjoying Bright Network?

My journey as a Silicon Engineer – Adam’s story

Book open Reading time: 3 mins

Meet Graduate Silicon Engineer, Adam from Graphcore who tells us about his journey to the firm, gives us insights into what he gets up to day-to-day and imparts his top application advice.

Tell us a little bit about yourself and your background?

I studied General Engineering at the University of Durham, before joining Graphcore as a Graduate Silicon Engineer in September 2020. At the start of my degree, I was sure I wanted to be a structural engineer (I still think bridges are cool!), but I soon caught the electronics bug which led to where I am today.

What stood out to you about Graphcore when applying for opportunities?

Graphcore stood out to me as an exciting place to join – offering an opportunity to work at the cutting edge of the fast-moving industry of AI computing.

On their website, I found the blogs and colleague experience articles gave a good impression of what life at Graphcore might be like. It seemed like a vibrant and cooperative environment – the perfect place to develop my skills in.

How did you find the application process for Graphcore?

I applied to the role in January 2020 and had completed the application process before the end of February. The process itself was straightforward and didn’t have unnecessary hoops to jump through.

After a short introductory phone call, the first interview stage was an online coding interview. This was a video call with an interviewer where I was given a short programming puzzle that didn’t last more than an hour.

The second, and final, stage was an in-person technical interview day at the Bristol office. It was engaging to spend the day discussing technical topics with a team of such friendly and driven people. I felt at ease in the office and knew, from that day, that this is a place I would want to work.

What is a ‘normal day’ like in your role?

Here at Graphcore, we create technology to accelerate machine learning. This includes developing our Colossus™ IPU processors, made with cutting-edge silicon processes. My role in this is that of a silicon verification engineer.

Silicon verification is the process of creating and running various tests to verify the correct functionality of the processor design before it’s sent for fabrication. Day-to-day, this mainly involves writing tests and models in C++, Python, and SystemVerilog. It’s quite a detailed and rigorous process but is very interesting as you must digest and understand the system specification and try to think of interesting corner cases where bugs may lurk in the design. Through this process, I enjoy the feeling that I’m developing a deep understanding of the inner workings of our IPU.

When issues arise, they’re often linked to interesting technical problems. It’s engaging and rewarding to collaborate with the other teams to find the root cause and satisfying to solve the puzzle.

As a graduate engineer, I appreciate how I had the opportunity to work on real problems from day one. Graduates here are integrated into the main team and I always feel my contributions were valued despite my relative inexperience. I’m only one year into my career and there are already elements of upcoming products that I can proudly say I made a mark on.

How would you describe the culture at Graphcore?

Everyone here is so friendly and supportive. On my first day, I was almost overwhelmed by the number of warm greetings I received from everybody (both in person and in my email inbox). Adding to that, the teams here are relatively small, which gives a nice feeling that you can get to know everybody quickly.

There is a strong culture of openness and trust within the company, where we can be trusted to produce our best work, while still feeling able to ask for feedback and support without judgement.

What piece of advice would you give to someone who is thinking of applying to Graphcore?

My top tip here would be to familiarise yourself with the basics of machine learning hardware and software. While I felt fairly confident in my general electronics knowledge, my engineering degree did not cover much in the way of AI. I would highly recommend a trip to the library if you’re in the same situation. Not only will it help with your application, but an understanding of the broader picture is essential to producing your best work.

Want to find out more about what Graphcore has to offer? Their profile is a great place to start – check it out here.