Two Years in the Tenure Track
Reflecting on my journey towards becoming a tenure-track astronomer
On January 3, 2022, one of my dreams came true: I began my first day as a tenure-track Assistant Astronomer at the Space Telescope Science Institute (STScI).1
Only about 10% of people with PhDs in Physics and Astronomy end up in tenure-track jobs.2 If so many other scientists have set their hopes on the same dream, then why is it that mine became a reality?
More than anything, I’ve benefited from privilege,3 good fortune, and being surrounded by wonderful people. Many factors for landing a tenure-track position were out of my control.
Nonetheless, I can share some thoughts about my journey so far, probably most useful for early career researchers (especially junior grad students). I’ll focus on the four following topics: (i) finding good mentors, (ii) making the most of serendipity, (iii) becoming an independent researcher, and (iv) staying happy and hopeful. Let’s unpack these below.
Mentorship
“There is no such thing as a self-made man. We are made up of thousands of others. Every one who has ever done a kind deed for us, or spoken one word of encouragement to us, has entered into the make-up of our character and of our thoughts, as well as our success.”
— George Matthew Adams
Many undergraduate students have asked me about where they should attend grad school. I always ask them back, “Which departments have supportive mentors and advisors? Are there at least two who you’d be interested in working with?”
Choosing excellent mentors is one of the most important factors for success, and I think this is true at every career stage.
Good mentorship can be subjective, by the way. Remember that you’re looking for the right fit for you—although it is still wise to check in with previous mentees about their experiences. Therefore you’ll need to truly understand your own needs and preferences. Here are some questions that you should ask yourself, especially when selecting a PhD supervisor and mentor:
Would you benefit from a hands-on or hands-off mentorship/supervision style?
What would does effective communication look like to you?
What would stretch you professionally?
How much freedom do you want to be able to explore your own ideas?
My PhD advisor, Andrew Baker, was pivotal for helping me develop both hard and soft skills. His array of multi-wavelength astronomy projects spanned multiple instrumentation modes, which gave me an incredible sampling of observational astronomy research. These research experiences broadened my perspective and kept me curious to learn more. Andrew also ensured that his students were well-rounded: we would regularly chat about work-life balance, written and oral communication styles, diversity and equity, mentoring others, and hobbies.4
I’ve also been spectacularly lucky to benefit from amazing mentorship at all career stages. My undergraduate senior research supervisor, Rachel Mandelbaum, gave me tremendous support during my first research project in astrophysics. My postdoc advisor, Josh Peek, showed me how my research fit in to the broader astronomical community, and how to strategically direct my efforts. These and many other mentors5 invested in me at different stages of my career, and I wouldn’t be here without them.
Serendipity
“I returned, and saw under the sun, that the race is not to the swift, nor the battle to the strong, neither yet bread to the wise, nor yet riches to men of understanding, nor yet favour to men of skill; but time and chance happeneth to them all.”
— Ecclesiastes 9:11-12 (King James Version)
Everybody knows that success in academia requires some degree of luck. This is clear if you scan the AAS Job Register, and compare this year’s hiring in specific subfields against previous years’ hiring campaigns. Sometimes, when you’re ready to hit the job market, there’s nobody hiring in your subfield! Or maybe your area of expertise happens to be loaded with other high-quality candidates who are seeking similar jobs. These factors are totally out of your control.
I used to have a graphic tee that said, “Make your own luck.” I believe that luck—good and bad—falls upon everyone, but those who recognize opportunity are best able to exploit it.
For example, I had been interested in machine learning since I was an undergrad, and I even did a summer internship in computer vision (but we basically used PCA for everything back in those days, so I wasn’t particularly convinced that we were anywhere near artificial intelligence). When I started grad school in 2013, deep learning had just burst onto the scene. However, I still didn’t think it was yet time to apply machine learning to my own research.
Why not? First, the software wasn’t mature enough. Tensorflow and Pytorch didn’t exist yet,6 and I didn’t think it was a good use of my time to code in a lower level language/framework like CUDA. Second, my research focus at the time was on relatively small data sets, which would have been a bad fit for machine learning. Third, I didn’t have a good enough understanding of astronomy fundamentals to convincingly publish an astronomical machine learning paper; as a first or second-year grad student, I was still learning the basics of stars, galaxies, cosmology, and the interstellar medium!7
By 2018, all that had changed: Pytorch was seeing widespread adoption among machine learning researchers, I had gained a much better understanding of galaxy evolution, and in a stroke of good luck, I had just heard about the Fast.ai course, an incredible deep learning course taught in a top-down approach designed to help you immediately apply new skills. But it took some investment to capitalize on this luck. Over the course of a few months, I worked on the Fast.ai courses during evenings, and I trained models on my gaming computer (yep, I used an NVIDIA RTX 780… how’s that for GPU poor). I also began asking around the department about ideas for machine learning projects (and GPU resources). Before long, I teamed up with a postdoc, Steven Boada (now in industry), got some idle GPU time from some physics professors,8 and we collaborated on a project that eventually got published!
This was just one serendipitous occasion, but I can identify quite a few other instances. In fact, there are numerous times that my entire career would have shifted had I not stumbled across one resource, or engaged in a conversation with one person.
I may not have even pursued astronomy were it not for the discovery of the Higgs Boson in 2012, back when I had just begun drafting up applications for a PhD in particle physics. This discovery meant the completion of the standard model of particle physics, and possible uncertainty in the funding landscape throughout all of particle physics. I wanted to be a part of the team that discovered the Higgs Boson, but it had just been done! Fortunately, there was an option to pivot into astrophysics: a few months earlier, I had heard that a new professor was joining the CMU Physics faculty, Rachel Mandelbaum. I had immediately reached out to say hello via email, which led to an offer to work on a research project with her, and thus began my foray into research astronomy.
I wanted to immerse myself in research before my first semester of grad school, so I started asking around the department at Rutgers for summer research projects.9 Although I had originally wanted to work on astrophysics simulations (i.e., theory), my only summer research option was to work on analyzing submillimeter and radio wavelength data—with Andrew Baker! He became my PhD supervisor and an invaluable mentor.
If Josh Peek hadn’t stopped by to check out my American Astronomical Society poster (which featured the machine learning project I mentioned above), then I probably wouldn’t have ended up doing a postdoc with him.
If my mentors hadn’t encouraged me to apply for the tenure-track Assistant Astronomer role in early 2021 (only 18 months after I completed my PhD), then I certainly wouldn’t have landed my current job!
A big part of life is navigating through circumstances outside our control, but sometimes opportunities fall in your lap. When serendipity appears, determination, alacrity, and hard work can help you take full advantage of it.
Independence
Let’s say you’ve made it most of the way through grad school. You’ve diligently followed your PhD supervisor’s advice, you’ve attended a few conferences, and you’ve published a paper (but you’ve got another one or two in preparation—just a few months away—I swear). How do you prepare yourself for an independent postdoc or eventual tenure-track research position?
First, you need to think about what you want to be known for. This is sometimes called a personal brand, although I’m not a fan of the term. Remember that you will be seen as your thesis advisor’s protégé, so you must distinguish yourself and prove to the world that you have your own good ideas. At the tenure-track level, you’ll want to convince hiring committees that you have a career’s worth of novel research direction ahead of you. Regularly set aside time to strategize, and think about whether your efforts support what you want to be known for. If you’re spending the bulk of your work time on activities that don’t enable your vision for the future, then ruthlessly prune them from your schedule.10
Second, you should broaden your network. I am also not a fan of the term networking, but there’s no denying that it’s important. Simply put, if you have a genuine interest in your work, then you shouldn’t hesitate to chat about it with others! Talk to people from other institutions. Email your peers when you see cool new papers on the arXiv. Ask senior professors about their work and insights. Offer to present on your work. Ask others to invite you to present on your work. Think about how new collaborations might shape your research program (but be very intentional here—dividing your time into many small roles in different collaborations can dilute your research efforts).
Third, you should curate your interests. Keep a collection of moonshot ideas and add entries when you head back from conferences. Stay on top of recent literature, and try to formulate your own understanding research trends in a long-term context (i.e., differentiating between fads versus research with lasting impact). Download random data sets or try new tutorials when you’re bored or feeling uninspired. Use social media if it helps. Social media can be a double-edged sword, but I’ve found several communities to be more beneficial than not: Twitter for astronomy and machine learning, Bluesky for astronomy, Reddit for machine learning, and Hacker News for machine learning and technology.
Fourth, establish yourself as a leader in your subfield. What does that look like right now? Ignore that researcher who published twelve first-author papers last year; what are other leaders in the field doing? What does it look like to lead the field five or ten years from now? Maybe you need to organize a conference and come up with a blueprint for your growing subdiscipline in astronomy. Or maybe you’re already the world expert in a specific method, but nobody knows about it, so you need to publicize it and go give a bunch of talks. Maybe you have strong but justified opinions on the state of your subfield, and you need to write a white paper. Maybe you’re making good progress and just need to continue doing good work and gritting out research papers. It’s your community—so shake off the imposter syndrome and help make that community better!
Fulfillment
He’s just a guy with a job.
— A priest at the Vatican Observatory, referring to Pope Francis, June 2014.
Remember, your job is still just a job. At the end of the day, you probably chose astronomy or academia or machine learning or whatever because it’s interesting and enjoyable. If you don’t enjoy it, and don’t think you ever will, then please do not feel obligated to continue! There are so many other joys in life.
If you do enjoy your work, then make every effort to sustain that delight. A healthy work-life balance is indispensable for preventing burnout. If you don’t set boundaries on your work, then it won’t hesitate to consume you.
One way to be satisfied with your work is to put it in proper perspective. There will always be more work and a higher position to attain. When I was an undergrad, I desperately wanted to get into grad school. When I started grad school, I desperately wanted to publish a paper (and guess what, I didn’t publish a first-author paper until my fifth year, so pulling those all-nighters in my first year probably didn’t pay off). When I was a senior grad student, ready to graduate, I wanted to land a good postdoc job on the East or West coast. And while a postdoc, I aspired to have a tenure-track position near Baltimore. Now what? I hope I can get tenure in a couple years, sure, but I’m no longer desperate to do so. I like the work that I’m doing, and I’m more interested in doing things that I find fulfilling and intellectually stimulating. If I were tenured, I’m sure I’d be looking forward to the next promotion, or becoming a division head, or getting elected to the National Academy of Sciences, or winning a Nobel prize, or whatever. But work is just one facet of life; don’t lose sight of that.
I can find gratification and fulfillment in other parts of my life when research isn’t going my way. Got a bad referee report? It’s okay, I still get to come home and cook and watch football and visit the aquarium with my family. I get to be alongside my toddler as she grows up and attends preschool and learns to ride a bike. I’m allowed to look forward to vacations and visiting friends. I personally hold on to the Christian faith, which helps remind me that my identity is not in my work. Other communities, relationships, sports, hobbies, and introspection can teach us these valuable lessons.
Life is a forest, so don’t let work—a singular tree—consume your view.
Final thoughts
There are many other things I could say, about imposter syndrome, honing your communication skills, creating a good CV and website, becoming a generalist versus a specialist, etc. I’m sure others have already covered those. In fact, the internet is full of advice, some of which is extremely high quality and applicable to all, and some of which is borderline harmful (or satirical, but these days, I can’t even tell). If you take in all this advice, filter it as needed, and add in your experiences, then I hope you’ll one day share with us your own set of lessons learned.
The Assistant Astronomer position at STScI is analogous to Assistant Professor at an R1 University, except that you support a NASA mission (or something similar) rather than teach courses.
I didn’t really research this statistic, but it comports with my anecdata, and seems to be supported by others’ as well, e.g., this Quora post.
One of the biggest advantages in academia is privilege. For example, tenure-track professors in the US are up to 25 times more likely to have a parent with a PhD (according to this Nature Human Behavior article). I am no different — my late father has a PhD from Purdue University, and I’ve benefited tremendously from his grit, sacrifice, determination, and wisdom. I am not only grateful for amazing opportunities that I’ve received, but also acutely aware of the challenges faced by underprivileged researchers.
Although I had stopped reading fiction in favor of non-fiction books for 10+ years, Andrew convinced me to pick up the hobby of reading fiction again. He argued that reading fiction serves not just as a way to relax and relieve stress, but also as a way to stimulate creativity, practice empathy, and inspire new writing.
I’m also immensely grateful to Kunal Ghosh, my supportive and kind undergraduate advisor at CMU, Marc Rafelski, who served as my postdoc mentor at STScI, and Laurent Pueyo, who is currently my faculty mentor at STScI.
Yeah, fine, Caffe and Theano also existed at the time, but I didn’t get the sense that they were seeing great adoption either.
In this case, good mentorship was also critical. My supervisor reminded me to focus on my first-year PhD project, rather than disperse my attention on trendy topics.
Thanks Matt Buckley and David Shih!
I recommend taking the summer off before grad school, unless you’re absolutely sure that you’ll be bored to death by not working on something. You probably won’t get another “summer break” in your life again.
It’s okay to have a vision that’s not fully clear, and therefore experiment with multiple avenues of research until you have a better picture. This was the case for me for most of grad school! But within a few months of starting my postdoc, I knew what I wanted to do with my career.