The Road goes ever on and on Down from the door where it began. Now far ahead the Road has gone, And I must follow, if I can… — J.R.R. Tolkien
This post is a work in progress — 2025 is still unfolding.
The PhD is done. The postdoc has begun. And something new is being built.
More soon.
]]>2024 was the year the PhD came to a close and the postdoc began. A year of finishing, defending, and pivoting.
Stay tuned.
]]>Here’s what I’ve learned.
Write down what you think your goals are. Not just for the PhD, but for the work. What are you actually trying to understand? Talk to people around you about those goals. Listen to what they say. Revisit it every month or two.
But don’t live at that altitude. The big picture is a compass, not a map.
The PhD is a long game, and long games will crush you if you try to measure progress at the scale of years. Each day, write down two or three specific, completable tasks. Do them. Check them off.
The daily wins compound. The feeling of progress — even small progress — is protective against the existential dread that visits every PhD student eventually.
The Type-A trap: looking at what the person at the next desk is doing and concluding you’re behind. You’re not behind. You’re on your own path, and the paths are not comparable.
This is easy to say and very hard to actually believe. I’m still working on it.
The PhD is, fundamentally, a training in tolerating not knowing things. You will spend years working on problems that might not have solutions, toward goals that might not be achievable. The sooner you make peace with uncertainty, the better your quality of life will be.
This doesn’t mean lowering your standards. It means learning to distinguish between the things you can control (your effort, your thinking, your writing) and the things you can’t (whether the experiment works, whether the reviewer is fair).
Every good researcher has a distinctive way of seeing. For me, it’s the basketball player’s obsession with systems, repetitions, and honest self-assessment. Find yours.
The PhD has a way of sanding off interesting edges if you let it. Don’t let it.
This sounds obvious. It isn’t. Exercise, sleep, eat food, maintain friendships outside of your department. You are not a research-generating machine. You’re a person doing research.
The two are not the same, and confusing them is the source of a lot of suffering.
None of this is original. The best PhD advice is the stuff your advisor won’t give you in a one-on-one: the honest, human stuff that takes a few years to learn. I hope some of this helps.
]]>EUGENe — the software toolkit I’d been developing for deep learning in regulatory genomics — was published in Nature Computational Science in 2023. That publication represented years of work: the idea, the design, the implementation, the experiments, the writing, the revision, the revision of the revision.
Seeing it in print was one of the more surreal moments of the PhD.
2023 involved more public-facing science than any previous year. Presenting at conferences, explaining your work to experts who’ve been in the field for decades, fielding tough questions — it’s a different kind of hard than writing. But it’s also energizing in a way that solo bench/desk work isn’t.
Preparing talks forces you to confront the question you’re usually too busy to ask: Why does this matter?
Not why it matters to the three reviewers who will read the paper. Why does it matter to the world? To a patient with Type 2 diabetes? To a clinician trying to understand why a variant is pathogenic? Answering that question clearly — to yourself first, then to an audience — is a skill I worked hard on in 2023.
Up next: 2024 →
]]>December 26th, 2022
Mornings are good for Coffee and contemplation Nothing else matters
Elixir of life Waking happiness of mine Another one soon
— Mom
Black or a latte? Each has its own pleasure A decision made
— Mom
Warm Honduran treat Dissipates my morning fog Better than birdsong
— Mom
Coffee oh coffee Coffee coffee my coffee Switching to decaf
More to come. The muse requires coffee.
]]>I have a tendency to try to learn everything at once, which means I often learn nothing deeply. 2022 was the year I started deliberately choosing depth over breadth.
There’s a difference between knowing about something and understanding it. Deep learning, regulatory genomics, the statistical mechanics of gene expression — I spent 2022 trying to actually understand the things I’d only previously known about.
Guitar has become a consistent thread through the PhD years. Some recordings from 2022:
Up next: 2023 — It’s showtime →
]]>What to do, as a machine learning expert, when the machine learning model is at odds with domain knowledge?
Embrace it.
— Christopher Molnar
One of the most honest things I’ve read about doing ML in a real scientific context. The tension between what the model says and what you know is not a problem to be resolved — it’s where the interesting science lives.
Brunson recently jokingly asked what Hart (career NBA scoring average of 9.8 PPG) does to earn his $12 million salary, and Hart said:
“Run around like an idiot during the game and just f— s— up.”
— Josh Hart
I think about this more than I should. There’s a version of every job — including research — that is just showing up with energy and making things happen by sheer effort and presence. Sometimes that’s enough. Sometimes that’s everything.
Tell me and I forget. Teach me and I remember. Involve me and I learn.
— Benjamin Franklin (probably)
I am not afraid of storms, for I am learning how to sail my ship.
— Louisa May Alcott
More added as I encounter them.
]]>Year three of a PhD has a particular flavor. You know enough to know how much you don’t know. You’ve been humbled by failed experiments, by papers you couldn’t understand, by reviewers who saw through arguments you thought were airtight.
And sometimes — more often than you’d like to admit — you find yourself reorganizing your Notion workspace for the third time instead of writing.
This is not unique to me. I think it’s a fairly universal PhD experience. The question is what you do with it.
I made real progress in 2021. My research direction sharpened, I got better at scientific writing, and I started contributing meaningfully to projects rather than just assisting.
I also attended AACR and ISMB, conferences that remind you that science is a community — that the papers you read were written by people who are just as confused and excited as you are.
2021 was a good year for music too. Some highlights from the sessions:
Does organizing your Notion count as research?
No. But knowing why you’re avoiding the real work — that’s useful data. 2021 taught me to be honest with myself about the difference between productive preparation and productive procrastination.
Up next: 2022 — I like to learn deeply →
]]>I joined my lab in 2020 and began the work of becoming a researcher in earnest. The difference between being a student and being a researcher is subtle but profound: as a student, someone else defines the problem and evaluates whether you’ve solved it. As a researcher, you have to figure out what question is worth asking — and that’s much, much harder.
The pandemic reshaped what a PhD looked like for my cohort. Labs closed, experiments paused, and the in-person community of science that I’d come to value was suddenly virtual. I leaned into computational work, which turned out to be formative — it cemented my identity as a computational biologist.
One of the things that kept me sane during 2020 was music. I play guitar, and the enforced isolation of the pandemic meant I practiced more than I had in years. Some recordings from that period:
There’s something about music that connects the left-brain/right-brain divide in a way nothing else does. Highly recommend having a creative outlet if you’re doing a PhD.
Despite everything, 2020 had its beautiful moments — trips to Michigan and New York when windows opened, reminders that the world was still out there.
Up next: 2021 — Does organizing my Notion count as research? →
]]>Every bioinformatics PhD student at UCSD goes through an intensive bootcamp at the start of the program. Statistics, programming, biology, algorithms — all compressed into a few weeks designed to level-set a cohort that comes from wildly different backgrounds.
I came in with a bioinformatics undergrad degree and two years of professional sports under my belt. I was neither the strongest coder nor the strongest biologist in the room. That was good. Uncomfortable, but good.
Your PhD cohort is everything. These are the people who will understand the specific frustration of a failed experiment at 11pm, who will celebrate with you when something finally works, and who will remind you that you’re not alone in the struggle.
Mine was exceptional. A group of curious, kind, ambitious people who I’m lucky to call my colleagues and friends.
The first semester of a PhD is less about mastering content and more about rebuilding your relationship with not knowing things. As an athlete, I was used to tracking progress — times, scores, weights. In science, progress is much harder to see.
Learning to be comfortable with sustained uncertainty was the real lesson of fall 2019.