About
Nice to see you here 🖐️
I'm Jérémy. I hold a PhD in analytical chemistry, specialised in mass spectrometry — most of my work orbits around LC-MS (liquid chromatography coupled to mass spectrometry): method development, quantification, and turning raw instrument data into something a lab can trust and act on.
For the past several years I've worked on the life sciences product side of analytical instrumentation — the people who don't just sell the box, but teach others how to get results out of it. That mix of science, instrumentation and teaching is where I'm most at home: I like explaining why a method works, not just how to click through it.
What I do
Day to day, my work is a blend of three things:
- Bridging instruments and software. Modern labs rely on a stack of lab informatics to turn measurements into usable knowledge. I help teams connect the instrument to the data workflow so results actually become decision-ready.
- Teaching and enabling. Running methods, training users, and translating between "what the chemistry needs" and "what the software can do." I write a lot of how-to material because a good procedure is one anyone can repeat.
- Tinkering with practical AI. I run local models on my own hardware for notes, search and memory, and I keep an eye on what's efficient and keeps data where it should stay. Privacy is not a footnote for me — it's the default.
The best instrument is the one a scientist can actually use. Software should reduce friction, not add it.
What I care about
A few threads run through everything I publish here:
- Doing more with less. Efficient models, self-hosted tools, lightweight setups that don't need a cloud bill or a data exit.
- Lab informatics that respects the scientist. The tool should serve the person at the bench.
- Writing things down clearly. If I had to re-learn it, a future reader should be able to as well.
Why this site
I built Jerem Flow as a personal space to keep that writing — the projects I tinker with, the notes I don't want to lose, and the occasional deeper dive.
A quiet sandbox: no analytics, no tracking, no pop-ups. Just Markdown in a Git repo that deploys when I push.
If something here is useful to you, that's the point.
Let's talk
I'm always happy to take the time to chat — about mass spectrometry, lab informatics, local AI, or whatever you're building. Don't hesitate to reach out 💬 by email or on LinkedIn:
See you around 👋