I was an early hire at a computational reproducibility startup for scientists [0]; the platform was basically an online frontend wrapped around a Docker container hosted on AWS, and the idea was that you'd put your code and data on the platform and have it be online-executable indefinitely, so you wouldn't have to worry about package updates, functions breaking, etc., because it was containerized.
The long-term goal was that scientists would describe their native software environments at a high level, and then the machine would build a Docker container that matched. In practice, your typical academic has no experience with containers/Linux/system-level dependencies. To prevent their walking away, I basically set up their software environments for them on an individualized basis when they reached out to us through intercom.
As PG says elsewhere, one of the main advantages of an early-stage startup is they can devote an insane level of attention to early users.
The long-term goal was that scientists would describe their native software environments at a high level, and then the machine would build a Docker container that matched. In practice, your typical academic has no experience with containers/Linux/system-level dependencies. To prevent their walking away, I basically set up their software environments for them on an individualized basis when they reached out to us through intercom.
As PG says elsewhere, one of the main advantages of an early-stage startup is they can devote an insane level of attention to early users.
[0] https://codeocean.com/