Real Time Analytics SaaS

Hacker News

Share card

Real Time Analytics SaaS

We have been involved in a number of real time streaming projects using tools such as Flink, Spark Streams and Kafka Streams backed by "real time" databases such as Druid. We always found these projects quite complex to develop and run, with stream processing in particular being a bit of a dark art. A stream to stream join in Flink can get quite mind bending for instance. In 2020 we had the idea of building a low code SaaS product for real time streaming analytics. The first attempt failed due to being a little over-engineered and with too many changes of direction, but over the last few months we have been building this with a simpler architecture and leaning on Clickhouse more. The SaaS is now live and we have a handful of customers in production and seeing the value. I would love feedback on the idea from the HN community and suggestions where to take the product next from a community that are likely to understand the value proposition and be familiar with the alternatives. Splunk is the main one that we get compared to so far. Website - http://timeflow.systems Walkthrough - https://www.youtube.com/watch?v=x4-WEHPt5-E A few blogs for our perspective and motivations: Build Vs Buy - https://timeflow.systems/build-vs-buy-for-real-time-event-streaming-platforms Why Is Stream Processing Hard - https://timeflow.systems/challenges-of-stream-processing-apps Why Move To Real Time - https://timeflow.systems/how-moving-from-batch-to-real-time-improves-the-customer-experience Thanks! Ben