I am interested in building a small product analytics tool that similar to Amplitude([login to view URL]) (a small part of that). Product analytics tools gets millions of events from different products (like xyz button clicked, xyz did purchased a product etc). And for a commercial multi-tenant products all these events and their internal data are quiet different and you'll get them in a JSON format. So we need to be able to query this json and allow people to slice and dice them. If you have already worked on a data visualization product like analytics tools, you'll know exactly what I mean.
From my research and my background I've chosen the following stack: React.js for the frontend server that use a modified version of superset ([login to view URL])
Node.js for the backend server
Kafka for streaming events
Druid for analytics DB ([login to view URL])
aws for hosting them
I'm looking for someone who has already worked on a similar project and especially has an expert knowledge in Druid and Superset. If you have experience, you should be able to answer the following questions:
1. How do you store JSON data in druid?
2. How does the frontend query look like?
3. If you have used Node.js to connect to druid, what npm module you used?
4. What else do we need for a product like this in production that can handle millions of events?
5. What was your role in the project?
If you answer them clearly and in details, I may hire you to set up mainly the Druid (and Superset) parts of the stack. Please don't apply if you are not specifically architected Druid DB.