 Hi, this is Yoho Sapil Bharti and we are here at KubeCon in Chicago and today we have with us Sudeep Goswami CRO at Traffic Labs. Sudeep is great to have you on the show. Great to be here. First of all, thanks for joining me today Before we talk about you know this event just remind our viewers. What is traffic labs all about? Yeah traffic labs was born as an open-source project Where it started off offering a reverse proxy and ingress controller and then started offering API Gateway And most recently we offer API management. We are here at KubeCon. Did you folks make an announcement here? We launched day 2 operations workflows today. When we look at day 2 operations Talk a bit about does it really end at day 2 and what about day 3, day 4, day 5, day 6? What I'm trying to understand is jokes apart. You said you know open-source is going to open source can solve day 1 problem Where you get everything you get it started day 2 is where the real challenge starts So just quickly talk about what does day 2 operation look like? What are the challenges and why you're focusing on that? There's day 0. There's day 1 day 2 Right, so day 0 is all about you putting something together a new infrastructure day 1 is really the first day when you go live and Day 2 is everything beyond that, right? So it's all encapsulated what happens once you have something in production And that's when we see all the cracks start to show in operations You start to run into scale issues. You start to run into like efficiency issues This is kind of what's encapsulated as day 2 and with traffic hub When we launched the you know a few months ago when we launched our first version of the product We wanted to get the foundation, right? So you can really build on top and create these day 2 workflows and we can talk more about what those are Yeah, I would like to learn more and also I want to look at day 2 operations from the perspective of API as well Exactly. Yeah, so if you look at what's happening In the world of API in a world of software development, but especially from an API standpoint I mean LLMs have created a step function in terms of the amount of code that's going to be generated, right? So when you look at API and you look at the velocity of development of new APIs that is going through a major transformation So it's become it has become a lot easier for now for developers to create an API and push it out there in the wild Right, but what that does is also creates a lot of issues around well This changes that you made are they the right changes? Are they going to break something in the environment? Are they going to work when you scale what happens when there is an incident? How do you know those APIs are working well before you have an incident? So all of these what we encapsulate as day 2 operational challenges They are not solved well and they create huge problems for our Customers who are the DevOps and the platform engineers. Let me look at Complexity of Kubernetes. It is there, you know people talk about it a lot Fact is that it is not going to go away. It might become even more complicated So some challenges arise from that complexity itself But a lot of challenges also arise from what a company is trying to do So so when we look at traffic hub and when we look at this, you know, the new announcement that you made What is specific challenge? Are you targeting the biggest challenge has been around? when you look at Kubernetes and when you also look at GitOps is To Unify everything to a common language Because if people are speaking different languages is going to be very hard for you to have this consistent unified experience and that is the first thing that we solved which was really making API management as code and In the world of Kubernetes everything becomes a YAML file. Everything is declarative And so that foundation has been missing in the industry. So that was the first thing that we did, right? So with that base foundation of API management as code now, let's look at it from a day 2 operations perspective, right? So the first thing is going to be Organizations are going to deal with change, right? Change is the name of the game now. So when a new change happens in the environment Because you have declared it as code what you can now do you can actually use the Git repository You can start using the GitOps workflow and one of the first new features that we introduced recently You know today one of the things that we do is when a PR request goes in when a pull request is You know initiated the first thing we'll be able to do is assess the impact of that change and do this Syntactic to semantic translation Which means to give some reasoning to what the change is impacting? So for example the developer thought the change was going impact two things in the environment Our analysis shows no you're actually impacting ten different things like that reasoning Unless you go through lines and lines of code is not going to be readily available to you So that's the first thing that we do now. Let's say you say, okay, this looks good. Let's move forward But then we have linters now that automatically run as GitHub actions Which are going to look for errors in the change that you made and give you the ability to just revert back the change So you're not gonna make create a breaking change the probability of you creating breaking changes goes down drastically with these changes Then let's say you deploy something for whatever reason there is an incident that happens We are able to offer Triaging so you can actually go back and see I made this change at 10 a.m At 2 p.m I started to see a problem as shown through my open telemetry metrics that My APIs are not performing the way they should so I can just go visually look at when was the last time a pull request was merged I can just go and undo that and very quickly see get it back to a stable state Before I can start troubleshooting it So you're minimizing the window that you have a downtime and you're increasing the you know the efficiency with which you're troubleshooting problems and how do they track incidents what They rely on different other incident management tools or you know traffic hubs Yeah, so we are not an incident management Tool right so they are using others. So if that says there is an incident They can use traffic hub to visually see and very quickly diagnose What was the source of that incident and be able to really minimize that quickly? So do you plug into that this image or it's totally something we would so what we do is you know We would we publish open telemetry metrics. So we have one of the most extensive open telemetry 20 plus Metrics 10 plus labels so gives you the slicing and dicing power and they can be exported to a third-party system like Grafana or Prometheus and Through there, you can actually see and reason your entire environment So our value is in the instrumentation that we provide the rich instrumentation that you can export to third-party systems If you look at the whole evolution of this, of course, there are a lot of things in your pipe And you cannot talk but just give us a glimpse teaser. Hey, these are the things that you folks are working on Yeah So I think we're I would say with all of this is exciting announcements that we launched today We're just at the starting point of day two operations because we believe this is the next frontier For a modern API management platform Anybody can come up with a bunch of features and put them out there, right? It is going to be about the operating model and how you're solving these painful day two challenges and that's the crux of the focus so what you just saw today is just a glimpse of what's possible and You know, you're gonna hear a lot more from us on this You know and then what we're trying to do if you think of it in like three pillars There's change management always going to be a big area of pain There's incident management always going to be a big area of change, you know pain And then there's what we call posture management is how do you have deep visibility and observability Inter API infrastructure and we're gonna focus on these three things because they really make up the day-to-day Operations of a DevOps or a platform engineer since we are here at a cube con which is open source event now Open source doesn't necessarily mean it depends on who you talk to right? Yes Source code should be open all those things are there But then there's another spec is that commercial support with that commercial support you're a stack at zero day They zero so talk a bit about the importance of open source for traffic hub at the same time talk about the importance of Commercial addition to support open source. First of all traffic lab started as an open source project Open source is very important to us. It is a core to the DNA however Open source by itself does not solve all of the issues So we when we first started the project we put out traffic proxy, which was our open source project We wanted to have a full functioning experience for developers Why and that's why we offered ingress controller and reverse proxy, which was a fully baked solution But we recognize the need that enterprises need more enterprise great features So what's going to be okay for a developer to do their job is not going to be okay for their managers or their directors and vps So that's why we introduced traffic enterprise, which was a paid offering which brought in high availability It brought in security and scale aspect to the open source project and then we released a traffic hub Which is also a paid product and what that does is builds on the capabilities of the API gateway or the traffic enterprise product and Brings in API management capabilities on top So really like you know the foundation was open source, but you can't just stop at open source You have to create more value for users So and you can start to address more and more of the use cases So Dave, thank you so much for taking time out today and of course talk about the new announcement here at cubecon And also kind of give us a glimpse of what to expect from traffic labs in the future Thanks for all those insights and I'll look to chat with you again. Thank you. Thank you. It's my pleasure