A Kaleidoscope of People: Dr. Adam Mepham

Welcome back to another edition of 'A Kaleidoscope of People', a series of posts that showcases the incredible individuals behind Kaleidoscope. For this conversation, Bogdan sat down for a patio coffee - on an unusually warm, February day - with Adam Mepham, in Washington, DC. We hope you enjoy!

Preamble: professional background

Adam has followed a winding path to reach his current position: a path guided by the dual threads of science and software. Although he focused on scientific pursuits during his postsecondary education, he was repeatedly drawn to the application of software to scientific problems. During his undergraduate degree in Chemical Biology at McMaster University, Adam’s senior thesis project entailed the use of computer simulations to describe the process of protein folding. And throughout his PhD in Biomedical Engineering at the University of Toronto, Adam regularly sought out ways to automate both data collection and analysis.

Following graduation from the University of Toronto, Adam spent a number of years working at Xorgate Solutions, a small technical firm, where he honed his skills in full-stack development. He now works as a software engineer at Kaleidoscope where he enjoys applying his scientific background to better understand the challenges faced by scientists and the best ways to solve them.

Welcome to another edition in our series. I’m here with Adam, sitting outside and having a coffee on what is an oddly warm day for February... I’d love to open with asking you how you transitioned from the biomedical space to software? And why?

Great to be here! So all throughout high school, I assumed I was going to be a scientist. For my undergrad, I did a Life Sciences degree, which started off general in year one, and then got more focused after that. I chose to go into Chemical Biology, which in a lot of ways is learning about small molecule drug discovery — it’s chemistry applied at biology, as opposed to Biochemistry, which is the chemistry of biology. In my fourth year, I did my thesis in a computational chemistry lab. Actually, I was working on a Python model for protein folding, back before this was a wildly popular thing in the media. I remember having to do a lot of work to understand the fundamentals of programming and build my knowledge from the ground up, and I really enjoyed it.

When I graduated, I started my PhD in biomedical engineering at a well known lab at the University of Toronto, and I just kept being drawn to software-related problems. We would be doing really manual analyses or visualizations, and I kept trying to find ways to build pipelines to automate this because it was so annoying to do in Excel. It was simple stuff early on, but it really appealed to me. And as I was getting closer to graduating and looking around at industry job prospects, I realized that beyond my own personal interests, knowing how to program was in high demand in the bio market, so I decided to commit to the path more formally. For me, this meant deepening and structuring my own learning of the space, and then looking for entry-level jobs that would let me do this more formally. Ultimately, I ended up getting a job as a software engineer at a small company, and building up a lot of competence working in the field over a couple of years. That’s around when Ahmed reached out about Kaleidoscope, and the rest is history.

As you weaved between chemistry, biology, and software, from undergrad through to the end of your PhD, what surprised you the most? What do you think is a really misunderstood aspect of science, that people who aren’t in the field might be surprised to know?

Honestly, how chaotic lab life can be. When you see a final, polished paper in Nature or something, you think that these are the most diligent, careful, calculated scientists who are doing this work. You only see the final product — you don't see that, behind the scenes, there are so many things that don't work, or that go wrong or break. Most often, those results get brushed away, and this sea of unstructured, chaotic work is not talked about, or the true details of the underlying methods are obscured, so the secret sauce of how to actually get the experiment to work is left out. I think this contributes a lot to the replication crisis we are faced with in science. It makes me think of what we could be doing to capture the process much more effectively… I mean, we walk around with powerful cameras in our pockets, what if we made people film their entire method? There’s actually a journal called the Journal of Video Experiments, where the requirement is that you submit a video of your method, which I love.

Yea, that’s really cool… on this topic of exciting things in the field, what’s something that you’re really optimistic about or love nerding out about, even knowing all the caveats you know as someone who has worked in the field?

The first thing that pops into my head is AlphaFold. I read the textbooks in undergrad, over 10 years ago: protein folding, unsolved problem, possibly unsolvable problem, one of the biggest problems in biology, all our models are terrible, we predict like 20 or 30% accurately, and so on. And then, in such a short window of time, we made huge leaps in this. AlphaFold 1 did 60 or 70% or something, AlphaFold 2 blew that out of the water. Obviously, it’s not a perfectly generalizable model, it’ll have its biases, it took a ton of carefully curated data, and so on, but this was a hard problem and we have to celebrate our successes. I think the rule of thumb is that it’s about one PhD’s worth of work to perfectly crystalize one protein, so for every protein there is a human who spent 4, 5, 6 years fighting for this… but you don’t see or appreciate that when reading about a model.

Love that framing, such an insane amount of work and coordinated effort and support that went into making something like this possible… speaking of support, what’s the best or worst piece of advice you’ve gotten from someone during your career?

It wasn’t advice per se, but my parents were incredibly supportive of me doing a PhD and then completely shelving that to move into software. A lot of people suffer from sunk cost fallacy and assume you have to keep doing what you started - I know I fell for this many times in my life- but my family helped me appreciate that it’s never too late to change and that I’m relatively young/early in my life, so remaining static doesn’t make sense.

Kind of the flip side to that is this idea that getting a PhD puts you head and shoulders above others when it comes to job prospects. I don’t agree with this, I think this is not good advice except for very specific fields, and not something I have found to be true as a broader rule. Lots of people do graduate school because it’s the ‘logical’ thing to do next, instead of pausing to really ask themselves why they are doing it and whether it makes sense. So maybe it’s actually not necessarily bad advice, but a lack of advice or open dialogue, and patterns of going down the path of (seemingly) least resistance, when in reality it’s a hard path and not optimal for most people.

Yea, especially back when you and I were students, there wasn’t as much of a normalization of scientists pausing to ask “wait, do I want to do graduate school?” before committing, as maybe there is today. Ok, hard change of direction here: what’s the last good book you read?

I used to read a lot, especially in high school. But I haven’t been able to set as much time aside for this over the years. I do still read occasionally, more for pure enjoyment, and fantasy is what I turn to for this. Brandon Sanderson is one of my favorites, he’s considered a master world builder and will do cool things like invent entirely new magic systems in each book. He’ll also weave different worlds together in the metaverse sense of things, and he has this one recurring background character who makes an appearance in every book, at critical junctures, no matter what world it’s set in, which is a cool detail. Oh, also, I read a sci-fi book recently called Children of Memory, which is a cool thought experiment on how different species could have evolved if their intelligence was supercharged/boosted rapidly across generations.

If you had to forever become an inhabitant of any fantasy world from any of the books you’ve read, what world would you pick?

There are a lot of interesting worlds out there… the Tolkien world is cool, I feel like there are a lot of normal people living normal lives there, but being a hobbit does seem like a sweet life. On the flip side, I do like that Brandon Sanderson creates these unique magic systems and introduces rules around, for example, who can learn how many types of magic. But the one character, Hoid, who passes between worlds, can pick up a skill from each world he passes through, so I would want to be a world hopper like him, in that multiverse.

Another fantasy-related question: I know you like playing D&D (I do too) — tell me about the character you created and are playing as!

His name is Garrett High Hill, and he’s a quippy, sarcastic Halfling who, honestly, is kind of a jerk. He and this wizard in our campaign like to get on each other’s nerves, which is always a fun time. Garrett is a Ranger, which is one of the simpler classes in D&D, but it’s nice to not have to spend hours before each game remembering what your character does. For Garrett, it’s simple: shoot at things and focus on building up accuracy. It’s become a bit of a running joke in our group because Garrett basically can’t miss any of his shots, so I almost don’t need to roll any dice. It’s a really fun game to play, and cool to see how our Dungeon Master (DM) decides to shape our experience each time we play.

Yea, I similarly feel super lucky to have the DM I have in the group I play with, really makes the game a special, unique experience. Ok, last question and one I like to ask our distributed team. You live in Toronto: imagine one of your friends visits you for a day, and you want to show them your favorite stuff to have a good time. What do you do?

There’s the classic but maybe more boring stuff, like the CN tower or Niagara Falls that people typically do. But what I would probably do is take them up to my parents’ cottage instead. This is a couple of hours north of Toronto, so it’s easy to get to, and it’s a really nice place that’s right on the lake, so there are a lot of fun things you can do together over the course of a cottage day. I’d maybe do a big breakfast to start, then go out on the boat or get pulled behind it on a tube. There’s a forest you can walk through during the day or you can go fishing if you enjoy that, and then I'd do food at the fire pit at night. The living room is also a really nice place to relax and read and stuff — the whole back wall is glass and looks out onto the lake, so it’s really beautiful.

That sounds really lovely and serene. Thanks for sharing that and for taking the time to chat today!

If you want to chat more about anything we wrote, or you’re interested in finding a way to work together, let us know!