I'm Andrea, a scientist turned data engineering leader.
My work has always been about connecting complex biomedical data and making it robust enough to build on, most recently in the knowledge graphs at BenevolentAI and BenchSci.
When I started my US job search, I applied the same discipline to AI. This site is part portfolio, part experiment. Its centrepiece is the pipeline that came out of it, and a demo of it you can try below. If you'd rather start with me, there's more on my background and experience.
I'm looking for my next role in California, USA — reach out onLinkedIn (opens in a new tab)
The project
The résumé-tailoring AI pipeline that isn't allowed to make things up
The demo takes a few minutes, uses fictional candidates and needs no sign-up.
Why I built it
Moving my job search from the UK to the US meant rewriting my CV for a different market, over and over. An LLM can do the rewording, but left alone it produces confident claims that aren't true. So the pipeline works from a record of my confirmed experience: it flags the gaps between a job and that record, asks me about each one, and only then drafts the résumé, with every bullet approved by me. Each answer stays in the record, so every résumé makes the next one better.
What it does
The demo runs the pipeline's steps in a small web app. Pick one of three fictional candidates and a job they're applying for, and you play the candidate. The AI matches the job against their CV, proposes skills to add, and asks about each gap: real experience they left off, or a genuine gap? Your answers decide what the tailored résumé may draw on, and you approve or edit every bullet before it's final. Every decision lands in an append-only audit trail you can inspect. All the data in it is synthetic.
Use it yourself
The pipeline the demo is based on is a public repo you can use for your own job search: the AI skills and the folders they work in, run with Claude or another AI assistant on your own private copy of the repo. You seed it from your own CV, résumé or a printout of your LinkedIn profile, and your data stays in your own files.
You don't need to be technical. My step-by-step guide is written for people who've never used GitHub: no coding and no command line, with example prompts to copy at every step.