- Valuation Model
- Expert Interviews
- Founders, funding
What is Shortwave's tech stack and how does it approach the decision of using third-party email APIs like Nylas versus building its own technology?
Jacob Wenger
Co-founder & CPO at Shortwave
Our tech stack at a high level is that our web and iOS apps are built in React and React Native, and on the backend, we're running on Google Cloud. Our backend is written in a language called Kotlin. We don't use any services like Nylas. Instead, we interact directly with the Gmail API. The main reason we do this is because there's a lot of interplay in Shortwave between the client and the server, more so than in most email clients. This gives us flexibility—interacting directly with Gmail gives us the flexibility to really tune that the way we want.
Some examples here. When you're talking with other users on Shortwave, we actually deliver your messages in real time via a side channel. You can see typing indicators and get messages immediately, and we still send it over Gmail on the backend. Being able to control the entire server aspect of this gives us the control to do both of these things at the same time in parallel.
Another example is that we do a ton of work taking HTML emails and, when possible, converting them into a rich text format that we've defined so that you get a consistent display experience. You get the same set of colors, you get the same text, you get the same formatting, and this is all possible because we’re actually doing all of this on our servers when emails are coming in. As far as build versus buy, I think there is a right time to do both.
We’re buying in a sense when, by using Gmail, we’re offloading a lot of the spam and the deliverability to them. We’re focusing on the areas we consider to be our core competencies and our core differentiators, which are the actual user experience and the workflow. The key here is don't outsource your key competency and focus on the areas that are the largest risk for your company.
The largest risk here for us is not necessarily that Gmail will cut off our access but rather that we won't build a differentiated product that actually achieves product-market fit. Whatever we can do to focus more time on that and less time on building infrastructure, the better.