Don't believe the hype: the product (a distributed in-memory database) isn't all that remarkable. There are many other contenders in a similar space (many open-source) with similar feature sets that are far more reliable and better stress-tested. Do your research before deciding to invest your talents here.
Of course, the product isn't terrible, so it's not surprising MemSQL has gained a small market share. Unfortunately, the company is terribly managed. This is most notable on the business side, where those in charge have surrounded themselves with sycophantic cheerleaders. As a result, they've created an environment where the agility so necessary to a successful startup is almost impossible. When fundamental problems arise, management will generally avoid the core issues and hire a new employee as a band-aid. This "growth" is then touted as a sign of health — when it is actually a result of dysfunction.
The problems with management are so obvious that they create a toxic internal environment. In my first week at MemSQL, I realized that a sizable fraction of my co-workers felt contempt for the company and were actively looking to leave ASAP. The turnover rate is astronomical (even for a startup). Not only is this detrimental to morale, but it is truly horrible for business — (as it takes over a year to train each new employee to expert status).
MemSQL does pay well, have a cool office, and offer various other fun perks. But keep in mind that this is because they have to — since the majority of their employees don't really want to be working there.
