In this episode of Mastering CS: Candid Leader Insights, Irina Cismas sits down with Viola Marku, Head of Customer Success at WunderGraph, an open source platform that helps engineering teams manage their APIs at scale. Viola joined WunderGraph about a year ago and was recently promoted from individual contributor to Head of CS, giving her a uniquely grounded perspective on what it takes to lead a small but highly technical team.
She shares how WunderGraph keeps its CS function lean without sacrificing quality, how the team uses AI to debug and reproduce issues in minutes, what changed when she moved from individual contributor to team lead, and why leading with curiosity is the one habit she carries from every role she has held.
What You’ll Learn
- What WunderGraph does and how a small but technical CS team is structured to serve enterprise clients
- How Viola balances staying hands-on with customers while leading the CS team
- What tools and processes keep a lean CS team productive without creating overhead
- How the team uses AI and Claude to debug issues, skip engineering back-and-forth, and serve more customers
- What changed when Viola moved from individual contributor to Head of CS
- What’s on the CS agenda for the rest of the year, including customer health scoring across all tiers
- Why leading with curiosity and getting to the why is the most important habit in technical CS
Key Insights & Takeaways
A lean team needs a do-it culture more than it needs tools. The biggest quality-of-life improvements at WunderGraph come from team members who spot a problem and fix it, often overnight. Culture drives that, not process.
AI works best when triangulated with human curiosity. Full AI automation fits self-serve products. In complex technical environments, you still need a human to understand what the customer is actually trying to achieve.
Debugging with AI skips 90% of engineering escalations. Using AI to reproduce and debug customer issues means CS can resolve most problems without pulling in engineering, saving time on both sides.
Maintaining the individual contributor hat is a strategic asset. Even as a team lead, staying close to customers keeps you aware of adoption patterns and what customers are actually saying.
Market your team’s work internally. The work CS does is the apex of the entire organization’s efforts. If you don’t resurface it, it goes unnoticed. Wins, lessons, and customer outcomes need visibility.
Don’t jump to fix before you understand the why. Fixing quickly without understanding the root need is expensive. The same feature request from two customers can mean completely different things.
Processes need to be sustainable, not just adopted. Many processes get picked up and dropped within a few quarters. The goal is to build habits that last as the team scales.
Podcast Transcript
Intro
Irina (0:04 – 0:41)
Welcome to Mastering CS Candid Leader Insights, the podcast where we dive into the world of customer success with industry leaders. I’m your host, Irina Cismas, and today I’m joined by Viola Marku, Head of Customer Success at WunderGraph, an open source platform that helps engineering teams manage their API at scale. Viola, I’m happy to have you here.
Thanks for joining!
Help me understand your world at WunderGraph today. What does the company do? What does CS look like?
And how is the team set up?
Viola (0:42 – 3:37)
WunderGraph is an open source platform to help teams manage federated APIs at enterprise scale, whether GraphQL, REST, gRPC, SOAP, and so on, in a single unified graph. We are deeply technical tooling for these teams and we support use cases at scale.
Some of our customers are the likes of eBay, who manage all their traffic through WunderGraph, which is an enormous amount of traffic. But we also operate with other industries like Royal Caribbean for their naval fleet and systems on board their ships. We work with entertainment companies like Paramount and FanDuel, and we also work with startups and scale-ups that offer different technologies to their end users.
We’ve been in the market for about four years now and we’ve evolved a lot over that time. I joined about a year ago because I had previous experience working with the founding team, and it was a great decision on my part to jump on board with WunderGraph.
In terms of the customer success team, we’re actually very small but mighty. The team is very focused and we operate specifically with senior customer success engineers. This is intentional because we operate in a deeply technical market and we need to be up to par with the technology we’re offering, all the feature sets and so on, and be ready to spot opportunities where a customer might want to adopt a certain feature set and do that with curiosity and intention.
We operate with a very small set of tooling, which actually serves us really well at the stage we’re at. As a startup, we’re still laying the foundation. The train is running really fast and we’re still putting the road up. But it’s a good approach to have a smaller team, a smaller set of tooling, less overhead in terms of processes, and just see what we need, see what serves us, and then scale from there.
Keeping the Team Lean: Tools, Culture, and the Balancing Act
Irina (3:38 – 4:52)
That’s the dream of any CEO: operating with a small team and a small toolset and trying to optimize the impact of the resources you have available, whether those are people, tools, or processes. There are so many follow-up questions from what you just shared. I also know that you were just promoted from CSM to Head of CS, so from individual contributor to having the overview of the team. How do you manage to keep the team lean and where do you put technology on top to help the team perform? What’s the thinking behind this formula?
Viola (4:53 – 7:28)
It’s definitely a balancing act, because the promotion didn’t make the individual contributor role fade away. A lot of my day still revolves around speaking with customers, speaking with users, making sure things are running smoothly, and interacting with the team constantly. We keep that communication open and we have the luxury to do so.
As a startup, it’s sometimes frowned upon that you have to get your hands dirty with so many different things, but actually that’s the beauty of it. You have much more ownership and much more ability to set the direction of things the way you perceive them to be more ideal. That’s definitely one positive.
How we balance setting up tools and processes is really about trying to see the cost versus the benefit of what we need to do. We need to make sure that whatever we bring on board, we can manage, we can set up, and that the investment will be used and will become part of our habit as a team.
Whatever we work on often comes from a pain point. That’s the biggest motivator, and it’s the same in a lot of the use cases we see with our own customers. If we see a huge pain point we’re experiencing, that’s typically where we naturally stand up and go fix it. That’s really the startup mentality: go where the fire is bigger and try to resolve it.
And the thing I’m probably most proud of right now as a team lead is the culture of the team, a do-it culture, a fix-it culture, where we’re putting together small but mighty improvements as we go along. Some of the team members I work with are based in the US, and sometimes I wake up in the morning and something has been improved overnight. It’s a fantastic feeling. The baseline of it is the culture, a team culture that wants to spot and fix things regardless of what the tooling looks like.
Irina (7:29 – 7:47)
And what helps the team to be productive? Because you mentioned tools, what are those tools and how do you combine it? What have you built?
What’s backstage to basically keep this format of lean team?
Viola (7:48 – 11:25)
We have been using a tool called Pylon for quite a while now. It helps us manage customer communication via tickets, email, or whichever avenue it’s coming from. Pylon is able to ingest it and assign it to the correct account. It also has a lot of AI features and automation available, and it really supports the routing of the team as well. For example, ticket assignment, there’s a ton of analytics we can leverage, and it’s also able to run NPS and CSAT scoring.
We’re currently running the NPS survey every quarter. We don’t go as granular as CSAT right now because we have quite a close relationship with all our customers. We’re in that fantastic golden era where NPS serves us really well combined with the other tooling we have.
Circleback, our AI note taker and call recorder, is also really good because it takes away a lot of the overhead of taking notes while also trying to manage a conversation. I’m still someone who takes post-it notes during calls, probably because that’s how I grew up learning. Writing things down slows the process a little and gives the brain more space to process. So the transcription tool is super important for us.
The great thing about Pylon is that we can integrate it with Circleback, which means all the calls we have with customers are ported in at the account level and update the conversation history and status of that customer. There’s also the ability to have custom notebooks per account, so we can create specific documents like product wishlists or really specific things we want to capture about a customer’s experience. That takes away a lot of overhead and helps us manage a very lean team without losing quality of delivery.
Other tools we use a lot are Slack and Teams for direct communication with customers. I think they’ve taken over email because they’re more conversational and instantaneous, and you can hop on a huddle with a customer very quickly. It feels like a much closer relationship than email-based support, especially now that we’re not traveling as much and not necessarily working with teams in our own regions. I work with teams in New Zealand, and that makes building a closer connection with customers more challenging. These tools really help bridge that gap.
How the Team Leverages AI in a Technical CS Environment
Irina (11:25 – 12:26)
Yes. I’m also curious, how do you leverage the AI? Because you work in a technical environment and you have a technical background.
And the latest conversations I’m having with senior CS people are feels like more like talking to an engineer than talking to a CS because now everybody is building AI agents for CS. Everybody is leveraging cloud code. Everybody is leveraging CGPT.
Not to summarize this way of thinking. I think that’s a new wave. So you being also with a technical background, I would assume, but I might be wrong, that you are also leveraging AI.
So I’m curious, how is an engineer person transformed into a CS leveraging the AI wave as we speak?
Viola (12:27 – 15:50)
We definitely are, and it’s also part of the success of the team and managing it with a leaner approach. I think that also insulates us from changes in the future that might affect how our team will develop and expand, so it takes a very measured approach there.
I personally think that a complete AI approach to customer success works specifically with self-serve tooling, tooling where you don’t necessarily need to have a relationship with your customers to make sure that they are leveraging it correctly, that they’re able to set it up correctly, that they’re making the most out of it. In complex technological products, we need to really triangulate the automation with a human, because that’s where we can still deploy human curiosity to understand what the customer is actually trying to achieve and what’s coming next, without losing sight of that human connection that brings us such a wealth of conversations, of insight, and a lot of the time helps us think out of the box.
Triangulating between the data, the pure mechanical answering, and the human touch is important. How we actually utilize AI at WunderGraph, specifically in the customer success team, is to help debug and reproduce issues very quickly using Codex. This allows a customer success engineer to quickly check if what the customer is reporting is something legitimate happening in the product or potentially a misconfiguration, and allows us to skip the back and forth with engineering teams in almost 90% of cases.
What that means is engineering has more time to actually do engineering, new product development, and innovate. And for us in customer success, it means a quicker response to the customer and supporting them faster. So it’s a win-win altogether. With Codex, we can actually even suggest fixes and speed that process alongside, which is a huge win for us. With Pylon, the AI features help us pre-craft some responses because it can be integrated with your codebase and documentation, which gives us a strong edge there.
In practice, we’re able to serve a wider number of customers within the same response time that we would have given with a smaller subset had we not had AI, maintaining high quality, maintaining a great NPS score, and managing a very workable amount of requests.
From Individual Contributor to Head of CS: What Actually Changed
Irina (15:52 – 16:33)
I want to switch from the company angle to you, and I want to understand how your world changed from the moment you switched from individual contributor to Head of CS. What was different? How did you manage to adapt? You work in a startup and you mentioned earlier that you can’t drop the individual contributor role entirely, so you still have to manage customers while also being the team lead. How do you combine those two worlds and manage to do two jobs in parallel?
Viola (16:34 – 19:11)
I think it’s potentially all down to the understanding that the cultural layer is what makes the team work cohesively. And ultimately, we can tackle really anything that comes our way as a cohesive team.
What changed for me was really having a different perspective on the team layer. I’m really happy and proud of all the people we have in the team. They work so beautifully together, they have such a great combination of characters, and they fit wonderfully within our ecosystem. That makes it a lot easier to sit down and have genuine conversations about customers, being transparent about where we stand, being clear about what’s happening, and taking on board their perspectives and approaches.
In practice, it means we really operate as a strong team, and what’s changed is really considering the collective versus the individual aspect. Focusing a little bit on the marketing of our team is also something that I’ve felt is often overlooked by leads and heads of teams. It means resurfacing the great work we’re doing, resurfacing the wins, and sometimes resurfacing the lessons as well, to make sure the work does not go unnoticed. The work we do in CS is really the apex of the whole organization’s efforts, and what we are doing is resurfacing it for everyone to see what ultimately hits the customer and how they make use of it.
I would also say that within the world of AI and the life-changing adoption we’re seeing across industries, maintaining your individual contributor hat is a great asset, because it really allows you to keep an ear to the ground on what’s happening and keep that sense of what the customer is saying or where a certain adoption pattern is heading. That’s a great strength to maintain.
What’s on the Agenda for the Rest of the Year
Irina (19:14 – 19:36)
What’s the biggest challenge you are trying to overcome, or what’s the thing that you want to fix until the end of the year? What’s your priority? Let’s not talk about the challenges or the struggles.
Tell me what’s on your agenda to try to improve until the end of the year? What’s your main focus now?
Viola (19:38 – 21:17)
With a growing team, we want to make sure that we have set up all the correct processes that serve us well at the stage we’re at. That means making sure that we have a strong customer health scoring that doesn’t just cover our enterprise customers, but covers all customers across tiers. That’s one of our big goals for the year, and of course it leverages the power of AI and automation, so it’s a completely achievable goal for us.
I also think it’s important that whatever process we take on and decide to run with, we do so in a sustainable way, so that it’s not one of those processes that I’ve seen many times in the past get adopted and then dropped after a few months or a few quarters, forgotten and taken over by other priorities. We need to always be introspective in our departments and be data-driven, but also reflect on our growth and the stages we’re at. In startups, you often don’t realize how quickly you scale in terms of your customer base, the number of tickets you’re handling, the number of engagements you have, and it’s important to reflect on that growth.
The Habit That Has Stayed Across Every Role
Irina (21:21 – 21:45)
You’ve been a UX researcher, a solution engineer, a strategic account manager, and now a head of CS, all in deeply technical companies. Is there a habit, a way of thinking, or something you learned the hard way early on that period that still shows up in how you run CS today?
Viola (21:46 – 23:47)
In my time as a UX researcher, I was deeply fascinated with the way we utilize software, from the interface down to human-computer interaction behaviors. I was really curious to understand how we developed over the last 40 years or so these UI elements that help us navigate tools and so on. That’s where it struck me that curiosity is really a big driver, really getting to the why. Why is it happening? Why are we trying to fix something? Why is the customer experiencing this problem?
I noticed over the years that there’s a tendency to jump in and fix quickly for the customer rather than stepping back and asking the why. Probably the biggest lesson has been making sure that the why is clearly understood, because it can be a very expensive mistake to jump in and fix without that clear answer first. You might speak with a customer who really wants a fix or a feature we’re not delivering, but perhaps another customer also wanted that particular feature in a slightly different way. Without really diving into their why, you would fix it just for one subset of customers. It’s a missed opportunity to look at the data, look at the information we have, and take all of that into consideration.
The biggest lesson is: lead with curiosity in all your conversations with customers. Get to the why in every interaction, be it explicit or implicit. Just get to that.
Irina (23:49 – 24:35)
Viola, this was an interesting conversation. I have to admit that I don’t often speak with the head of CS in one month into the role, so I don’t get the chance of basically knowing what’s happening and what’s behind the curtains when you are in the first month of the mandate. And I generally want to say thank you for accepting the invite and for speaking so openly about what’s happening in the current CS team that you are basically managing.
To everyone listening, thanks for tuning in. Until next time, stay curious, keep learning, and mastering customer success.
