scale.bythebay.io: Sanjeev Kulkarni Interview
Recording: scale.bythebay.io: Sanjeev Kulkarni Interview
My name is Sanjeev Kulkarni. I work at Streamlio. I'm one of the co-founders of Streamlio. The key engineering issues that we run to work is twofold: scale and simplicity. Uh people want to be able to store and analyze tons of data, petabytes, uh whatever, uh and you know, historical as well as real-time, so kind of variety of data. And at the same time, they want to be able to do this very simply, without having to have a lot of friction, without having to uh yeah, cause a lot of pain in terms of developer productivity or operational uh productivity. So, that's what we address. Uh one one simple stuff is uh there is no solution There is no messaging solution today that's available that's enterprise-wide
You know, people cannot Current messaging solutions stop out at like maybe thousands of topics or tens of thousands of topics. And if you really want to have millions of topics, you are looking at tens or hundreds of clusters and the management hurdles all the management headaches associated with that. With ours uh with our architecture, it's just one seamless across the firm uh subsystem uh that is so easy to manage and that that scales and analyzes everyone's data. Uh um yeah, so uh I I think one of the biggest uh challenges that's going to be there in the data industry in particular is the metadata. Uh right, people are consuming various data sets and and generating various data sets. So, a lot of the times in the enterprises there is no place to actually find what kind of metadata what kind of data is is available. Essentially the metadata of the data that's there is not findable in a single thing and and that is increasingly going to be a problem for us. So as as people's needs grow and and as people start doing all sorts of things, these kind of things become important
Uh functional programming number one has kind of significantly reduced the barrier of entry for a lot of analysis. So previously just to give you an example, we we we support a bunch of different kind of API layers. The low level API that we have is too low level. You have to stitch in a lot of things and and and sort of have to understand a lot of system nuances to kind of make things work. With functional API, what you have is a seamless way of expressing your computation and the system kind of takes care of stitching things for you. Basically understanding the nuances. So this means now that more and more people have access to the APIs and actually able to do useful things with it. Uh one of the most favorite parts about Scale By The Bay is that I come back to Twitter headquarters, my alma mater in some sense
Um so it's fun to see familiar faces. It's fun to see familiar settings um and fun to meet all the people.