Opinions · Career Development
Career Development
Notes from my own career so far, for students and early-career researchers. I am still early in my career myself, so I welcome advice: please write to me.
What I learned applying for postdocs
If you are applying for postdocs, especially with an unusual path or a publication list that looks scattered, these notes from my own search may help. I studied French literature as an undergraduate, spent a year on computer science, and then did a PhD in physics that ranged from stars to cosmology. When I applied for postdocs, I worried that my papers looked like disconnected pieces, that my background needed explaining, and that I would sound overconfident about things I had only started to learn. Here is what helped me. Every search is different, so take what fits yours.
A scattered record is a story you choose
- Tell a different story to each kind of position. For a project postdoc, lead with what the group needs and leave the rest to your CV. For a fellowship, embrace the breadth and propose projects that join the pieces.
- Pick one backbone for most of your fellowship applications. Mine was image processing for the Roman Space Telescope.
- Tell an unusual background in a sentence or two, as part of the story rather than something to hide.
Read beyond the ad
- Before drafting, read about the people: their recent papers, websites, and any documents attached to the ad, such as a funded proposal or a role profile. They show what the group really needs.
- Bring your own idea of where you would contribute, then test it against what you read.
- Name collaborators only where the connection is real. If you have met someone only briefly, connect through shared science, not shared events.
Fellowships
- If you build tools, show the science they unlock. A proposal is about the future and should show your own ideas.
- Two or three linked projects can span risk: a safe extension of your thesis first, then something more ambitious. The discussion sections of your own papers are a good source.
- Read the current call and its rubric yourself, and treat page limits as exact.
- Read past successful applications for their structure, not to compare yourself, and give this reading a fixed amount of time. You are seeing polished final versions, not drafts.
- Choose hosts you have real working ties with.
Protect your letter writers' time
- Letter writers are busy and may not be able to tailor every letter, so do the tailoring in your own cover letter.
- Give each writer one packet (CV, a core proposal, the list of positions) and send one bundled reminder a week rather than one per job.
- Keep track of which jobs ask only for names and which need letters.
Be precise about what you did
- Claim only what you can explain in an interview. Say who did the work: "I" when it was you, "we" when it was the team.
- Keep every document consistent: paper statuses, numbers and titles drift between versions unless you check them.
- Leave out what the ad does not ask for, and keep the cover letter and the statements from repeating each other.
Using AI without being misled by it
What I learned about using an AI chatbot for applications is on Working with AI.
Staying organized
- Keep a tracker, and check the job boards on a fixed rhythm, looking only at new posts each time.
- Have a core application (CV and general statements) ready early, and tailor it for each deadline.
- Prepare a long talk, a short talk and a two-minute pitch before invitations arrive, and an answer to the hard question each job is likely to ask.
Looking after yourself
- Comparing yourself with others is hard to avoid. Limit it, then go back to your own story.
- Talk with others on the market, and remember who you are outside it.
- Focus on what you control, and mark each submission done.
Reading list
Resources I found useful.