I used to be a regular backend developer, then I started a biology PhD. It seems like lots of people do computational work, but the landscape of devtools is pretty sparse. This is slowly changing (partly due to new AI-for-science interest), but it feels like there's still lots of low-hanging fruit to be built.

E.g. data infrastructure, declarative tooling for experiments, visualisation tooling, data sharing/exchange services, provenance infrastructure for threading data generation through to ML model training, protocols for lab equipment, etc.

So I'm wondering if people on the more traditional side of tech are interested in building devtools/infra for science, what you see as the main barriers, whether there's anything specific you'd like to build, and what would help you get started?

For those working in science, I'm also curious what tools you'd like to see built, and whether you'd be willing to collaborate with an engineer to make them happen?