My Experience

Sometimes your life only makes sense if you're the one living it. When I talk about working in ocean conservation, people sometimes assume I'm a marine biologist or have a similar background. While that would be very cool, I am not a marine biologist. I'm a former astrophysicist. I worked in quantum computing for a few years, and now I've made my way into the conservation world. I know exactly how disjointed it sounds. But it really is a fairly straight trajectory, and this sort of thing is not uncommon for former academics.

As an undergrad deciding on grad school, I was under the impression there was reasonable demand for astrophysicists in this world. I thought the degree itself was the hard part, and the rest would fall into place. By the end of my PhD, I understood this was not true; only a shrinking minority of PhD students and postdocs trickle into the occasional open faculty and research positions. I realised that, unless I lowered my standards a whole awful lot, astrophysics would just be a part of a longer journey for me. So when it was time to defend my thesis, I came to the fork in the road that most academics do. I sort-of wanted to stay in academia, but I didn't want to compromise on location or having a livable salary for the first time in my life. This made the sparse roles available much sparser. I applied to maybe 6 postdocs in places like Scotland where I could really see living, but ultimately nothing panned out. When the last one slipped from my fingers I gave myself a day or two to lament the loss of a romanticised future at a snowy observatory I’d cooked up for myself at age 19. Then I refocused. I would rather move on than live in a place I hated getting paid barely enough to keep going.

My PhD research was on angry little stars called M dwarfs. I was particularly interested in the impacts their volatile activity has on the planets around them, and the ability of those planets to develop/sustain the right conditions for life. tl;dr on my thesis, a lot of these planets are cooked. I doubt we're alone, but I do think it's a lonely universe where the delicate conditions needed for life are very rare. I was spending so much of my time thinking and writing about this, while watching the worsening wildfires and biodiversity collapse on our own planet. I couldn't (can't) stop thinking about how incredibly lucky we are to have the planet we do. I realised I would like to do something to help protect it, even if it couldn't be my first step out of my PhD. And of course it wasn't.

So why start with quantum computing? I'd like to give you a cute little story about how much thought I put into it. But really the role just kind of fell into my lap the week I graduated. I worked out my interest in the field later. Quantum computers have gone through periods of hype, and they were at a peak at the time. I think the hype has calmed, because once again we are 6 years beyond the last wave of quantum startups claiming we're 5 years away from quantum supremacy. I knew the field was saturated with grift, but there is also genuine promise and incredible work being done with quantum computers. I was excited by the possibility that they could radically reduce the power needed for advanced computation. Energy-intensive gen AI could cut down its footprint, and I was into that. However, progress in quantum computing is slow and they are far from delivering on that hope. And for my work they were essentially unusable.

If you don't know what a quantum computer is, it’s like a regular computer but worse. That’s an unfair characterisation, but for the machine learning work I was doing, not untrue. Mostly I ended up working with classical computers because quantum computers just couldn’t do the things I needed them to do. So, really, I would say my time with quantum algorithms was really just me being a regular machine learning scientist.

Landing my first few jobs in industry startups was relatively easy (though the jobs themselves were very much not). Making it into the conservation space was harder. I spent forever applying, interviewing, writing cold emails, talking to whoever I could, and fastidiously checking climate job boards before I broke in. I spent some time working with a pre-funding stage climate startup; however, while I’m not easily motivated by lining my pockets, I can only go so long without a paycheck. After all this effort, things worked out well for me and I very happily arrived where I am today at a nonprofit. I am doing work that makes me hopeful.

The underlying subjects in my journey - stellar astrophysics, quantum computing, climate law and practices, ocean conservation - are all very different. But the skillset is similar. It isn’t a big leap from coding up stellar models to coding up quantum machine learning models to coding up regular machine learning models. Really a fairly straight trajectory.

The job market in tech was tough when I came in, and it’s worse now. I really sympathise with the academics struggling to break into industry today. If you’re reading this, I imagine that’s you, and you’re here because you’re hoping I can help. I wrote this because I want to. The things listed below are what helped me and/or people I know in a similar position the most. I hope some of them are useful to you.

1. You already have most of the base skills you’ll need, but you should fill in gaps in your knowledge.
When I first started thinking about industry, data science and machine learning were the most obvious options (and most of my advice here is tailored to them). While I was very familiar with statistics, had modelled stellar activity, and had done basic machine learning, there were still a lot of gaps in what I knew. I supplemented this by following online tutorials and taking a free course.
I used Kaggle and TowardsDataScience for tutorials/basic info. The exact course I took is probably outdated; it was Andrew Ng’s free machine learning course, but I recommend finding another similar updated course. I know that MIT has a free one.

3. Tailor your resume but don’t spend forever on it.
One simple thing I did was create a few resumes tailored to the genre of role, each focusing on the aspects of my experience that best matched that genre, and likewise with cover letters. For example, in the first steps of my journey when I worked in quantum computing, I made resumes that emphasised my physics background. However, when I was applying to conservation roles, I stressed the applications of my work and my personal experience with climate organising in my community.
There are tools you can use like Canva to make aesthetic and creative resumes, and I highly discourage using them. Many companies use automated software to ingest resumes, and if your resume format is non-standard, it may get thrown out. There are some apps you can find online that will rate the “readability” of your resume and give you suggestions on how to improve. They can be helpful, but I recommend removing your personal information from your resume before uploading.
I didn’t do this next one myself and don’t necessarily condone it, but the job market is a jungle and people are desperate. A lot of places get a flood of resumes and use LLMs instead of an actual person who can pick up on nuance to sort through them.
I know that some people are putting their resume and the job posting into an LLM. They’ll ask the LLM to reframe their resume so that they are a good fit for the role. Again, I didn’t do this so can’t speak to its effectiveness, but if you do decide to try it, double and triple check the modified resume to make sure the model didn’t make things up.

5. Honestly, use flash cards.
This one is in the same vein as the last one. You can be super familiar with a concept, just as you are super familiar with your experience, but putting it in succinct interview speak is hard. If you asked me to explain backpropagation from scratch 2 years ago I would've exploded. Have a good, concise answer to common concepts in ML.

7. Get caught up on Python and SQL.
Thankfully I used python throughout my PhD, but I did need to learn SQL on the fly. Python is the most frequently used language in most roles that I’ve seen, although occasionally R shows up as well. I’m of the once-you-know-one-you-pretty-much-know-all mindset to common coding languages, but if you aren’t familiar with Python definitely take the time to familiarise yourself as much as you can. I’d also recommend reading up on distributing computing if you don’t have any experience with it, and maybe do a trial with AWS or Google Cloud so you really know what you’re doing.

9. Network.
Try to find events, virtual or in-person, where you can meet people working in industry. A lot of cities have tech meetup groups that meet regularly, and conferences will typically have job fairs or similar.
If you’re focused on finding a job that aligns with a specific field (like myself with conservation and climate work), specifically seek out that community in your city. Whether it actually helps you land that role directly, it deepens your connection to your mission.

11. Check on a number of job boards, not just LinkedIn.
LinkedIn is the primary one I see tech jobs posted on, but other boards like Indeed or Ziprecruiter can be useful as well (I don’t recommend paying for one though!). Many fields will also have job boards specific to that field, so it’s a good idea to seek those out as well.

I was methodical in my job search, and the above were the most useful steps I took to transition out of academia. I hope some of this has been helpful to you. Good luck in your search; and don’t lose hope or confidence. The market is hard, and landing a job can take a long time.