Not a side project for us[0], but lack of community beforehand/notoriety. Some companies with backing or from an accelerator/incubator are built around what is

Not a side project for us[0], but lack of community beforehand/notoriety. Some companies with backing or from an accelerator/incubator are built around what is basically one feature in our product, but they have more engagement/notoriety because they're branded. It's interesting to see that dynamic, especially given the fact they don't have a background in the problem/domain and we think the problem they're solving is not what actually matters. However: given the nature of our activity [we build ML products for enterprise] and have been profitable for the get go with this activity [no need for VC money], we're building the product for ourselves. We made a decision to develop that platform as if it were a normal SaaS to put positive pressure on ourselves and round off rough edges. For example, early users of the platform were important in that they helped us focus on common/irritating quirks we were aware of, and expose problems we were not aware of. One issue we were aware of is the "tragedy of the commons". Users were training models and competing with each other for resources. We then quickly made users able to schedule notebooks because we were hitting that problem often and it was really annoying. Another problem is troubleshooting code for users. We added near real-time notebook editing. Around thirty early users were students of one of our colleagues who used our platform in their machine learning final year project, and given COVID and financial resources, being unable to set up compute environments, we onboarded them and fixed whatever they were complaining about. - [0]: https://iko.ai https://iko.ai