You shouldn’t send the same résumé to every job. But you also shouldn’t spend three hours rewriting it from scratch for every new opening.
The trick is a strong base résumé, then a few deliberate changes per application. You don’t need AI to do it — you need to know what to look for.
Start with the job description
Read it before you touch your résumé. Look for three things: what will you actually be doing, what skills are they asking for, and what seems most important to the employer?
Most job descriptions mix essential requirements, nice-to-haves, and generic HR copy. Find the parts that describe the actual work — those are the things your résumé should make easy to find.
Don’t invent experience. Reorder it.
Tailoring doesn’t mean adding every keyword. It means changing which parts of your experience get the most attention.
If you’ve built an ML model, done statistical analysis for a research project, and built a web app — the research role gets the analysis near the top, the ML role gets the model, and the software role gets the web app. Nothing about your experience changed. You simply changed which story you were telling.
Match skills to evidence
Don’t just write:
Python, PyTorch, NLP, data analysis
Write something that gives those skills context:
Built and evaluated a PyTorch model for an NLP research project, using Python for data preparation, experimentation, and analysis.
This is especially important for researchers and recent graduates who have plenty of technical experience but limited conventional employment history. A project can demonstrate a capability just as well as a job can.
Keep a library of your best descriptions
You probably don’t need a different résumé for every employer — you need a library of strong descriptions for existing experience.
The same project can read differently depending on the role:
- Data science: Analysed a large text dataset to identify patterns and evaluate model performance.
- NLP: Developed and evaluated an NLP pipeline for low-resource language data.
- Research: Designed experiments to evaluate competing modelling approaches and investigated failure cases.
Same project. Different emphasis. Much faster than starting from scratch.
Keep a master résumé
Don’t overwrite your only résumé when you apply. Keep one comprehensive version as your source of truth, then create a shorter tailored version from it for each application.
That makes experimenting much easier — you can remove a project without worrying you’ll forget it, and rewrite a bullet without destroying the original.
Don’t let the résumé carry everything
There’s a limit to how much context fits in two pages. “Developed a transformer-based model for low-resource language glossing” is useful — but it can’t explain the research question, methodology, experiments, and results without becoming a thesis.
That’s what your portfolio is for. With Takeoff, those projects have a permanent home and you can link to them from any application. Your résumé stays focused. Your portfolio provides the depth.
Tailoring without AI is faster than you think
Once you have a strong base résumé and a library of your work:
- Read the role and identify the important requirements
- Pick the experience that proves them
- Reorder and rewrite a few bullets
- Check your portfolio link
No prompt engineering. No generated buzzwords. Just your experience, presented in the most relevant way.
Your experience doesn’t need to change for every application. Your emphasis does.
Takeoff gives you a place to keep the work behind your résumé.