She started her first tech job over 15 years ago as an embedded QA tester for a video game studio and has extensive experience in AAA game development, tools engineering, and AI. With roots in Ukraine and now residing in Stockholm, Sweden along with her two cats, Liza Shulyayeva has recently joined the QA.tech team as a Staff Software Engineer. Not only does she have an impressive CV, she also writes science fiction in her spare time, grows roses and lifts weights.
Sci-Fi, gaming and test automation
Sometimes, things just line up and opportunities appear that are exactly what you’ve been looking for. Liza teaming up with QA.tech was one of those things.
“After my last contract ended I took some time to think through what I really wanted to do next. I then decided I was especially interested in a specific subset of AI development: autonomous agents that run in fairly permissive environments with capability to reflect on and modify their own performance,” Liza explains.
And it seems the stars were aligned. With QA, test automation, and AI experience in her CV, it’s close to the perfect match.
“When I saw the QA.tech position and talked to Hanadi, it sounded like exactly the kind of thing I’d just resolved to keep an eye out for. After talking to the team about their vision for QA.tech, the decision to join was a no-brainer.”
After spending several years in the gaming industry as a tools, core tech, and backend engineer, Liza transitioned to working with WebRTC-based video APIs. Her focus included developing tools to integrate real-time and post-processing video and audio workflows with AI applications. These applications range from disfluency removal and conversational querying of large video libraries to real-time meeting assistants, among other functionalities.
But her interest in programming began much earlier, sparked by a fascination with video games.
“One of my favorite childhood games was Creatures 3, which sparked my interest in virtual life and led to several years of work on a snail life simulation in my spare time. It was a mess that was never finished—but a really fun mess to make!”
Apart from video games and programming, Liza also writes science fiction and has one bestseller on Amazon.
What else do you do outside of work?
“Besides writing science fiction, I grow roses on my balcony, cuddle my two giant cats, and drink coffee for people who hate themselves and know their life has no real purpose or meaning.”
What’s your impression so far of the team and the culture at QA.tech?
“It’s only been a week so who knows, someone could still decide to bring surströmming for lunch. But so far so good! Everyone’s been really welcoming and helpful, and for an early-stage startup I was pleasantly surprised by the really thorough READMEs to get up and running with the monorepo,” Liza says.
“I believe self-reflecting autonomous agents is one of the really exciting applications of AI to explore.”
“We’re at the tip of the iceberg”
With seniority in software engineering and experience in AI, Liza will be an invaluable asset in QA.tech’s mission of leveraging AI to blend the expertise of QA engineers with the efficiency of automated testing. We’re eager to learn what drives her and what challenges she anticipates in her new role.
What part of your job are you looking forward to the most?
“I’m really looking forward to helping build and validate a real-world, advanced use case for AI. ‘AI’ in general has become such a buzzword; there are many grand ideas that don’t always translate into something concretely useful. I believe self-reflecting autonomous agents is one of the really exciting applications of AI to explore. QA.tech’s testing agent is a perfect opportunity to build something really practical with this technology while pushing the limits of what we can do today. But in a more hand-wavy sense, I also look forward to having a tiny part in helping shape the future of autonomous agent applications. I think we’re at the tip of the iceberg for what we can do in this area, and the exploration and problem solving we do now can unlock huge potential going forward.”
Let’s dive into the AI thing a bit more. As you mentioned, QA.tech is an AI-powered platform. What is your personal view on AI?
“I believe that AI will be a net positive for the world if we do it right. Like other powerful technologies, it will require change and adaptation at both individual and societal levels. As developers working with AI, I think it’s our responsibility to tread thoughtfully and consider the potential impacts this work will have on those around us. This is why I see both the optimism and pushback to AI as a positive thing in the long run. I believe actively challenging and regulating what we’re building, just as with any other powerful technology, is an important part of building something as robust, useful, and ethically sound as possible.”
What do you think will be the biggest challenges?
“I believe the biggest challenge will be creating an agent that is flexible and ‘intuitive’ enough to act as a human tester would, but also constrained enough to produce consistently stable and reliable results. The flexibility of LLM output is part of what makes it so powerful, but it also creates the need to ‘wrangle’ models to keep them within reasonable guardrails. Compounding this challenge is the variance in performance and underlying configurations of the third-party AI APIs we often rely on! For example, consuming developers often don’t get great visibility into what OpenAI is changing under the hood. We see changes in GPT-x response speed and quality on a regular basis, but often can’t quite put our finger on why or what changed exactly. I think part of solving this challenge will be finding a good balance between relying on third parties and hosting cost-effective custom models locally for more predictable configurations.”
While we see more women exploring a career in tech and development today than 10-15 years ago, there’s still work to do in this area. Do you have any advice for other women aspiring to work in tech?
“Probably just to always remember what it is you’re excited about doing in tech and focus on that thing. There are many distractions, especially in online spaces. Some of them are unavoidable, but in general wasting energy on peripheral noise is only going to keep you from the things you were excited about in the first place. So as much as possible, try to preserve your time and energy and focus on all the awesome stuff you wanted to build. And if a job listing resonates with you and you think you’d be good at it, apply even if you don’t meet every single bullet point on the criteria list.”
