Stirring innovation
A power blend of teamwork, data and AI (part 1)
Agile by design
Agile by design
Meet Simone, Laurent, Zhilan and Vito – a team for whom agility means rapid prototyping, continuous learning and a relentless drive to deliver new tools and services. All while staying ahead of regulatory demands and pushing the boundaries of what’s possible with data and AI.
How do they make it happen? Let's find out.
Tell us something about yourself and your roles.
Laurent: I’m a Data Engineer in the Data Analytics Pod, that is part of the Trade & Transfer Assets (TTA) stream, based in Zurich. Over the past eight years at UBS, I’ve focused on data warehousing, modeling, reporting and analytics. One of my early contributions was helping build our Big Data platform called DARTH (it’s Data Analytics & Reporting Technological Hub, but yes, we love Star Wars references). Before UBS, I spent nearly two decades as a database consultant, mostly at Credit Suisse.
Vito: I’m also based in Zurich and part of the Trading Analytics Team within GWM F2B Digitalization division. I joined UBS over a year ago, after completing my master’s at ETH Zurich, where I specialized in Machine Learning and Deep Learning applications in engineering. My thesis led to a publication, which I’m very proud of.
Zhilan: Hi! I’m in the same team as Laurent but based in Singapore. I joined UBS through the Graduate Talent Program over a year ago, and so far, it’s been a rewarding journey of growth and exploration.
Simone: And I lead both teams – the Trading Analytics Team, which acts as the quant hub for our Investment Bank’s Execution Hub trading business in Switzerland and APAC, and the Data Analytics Pod within the TTA stream, which is part of our tech organization. Our operating hubs are in Zurich, Singapore, Wroclaw, and Pune. My background spans trading, asset management and data science, but always rooted in banking.
You’re all about practicing with purpose: prototyping, testing and experimenting. Can you share some details?
Vito: This has been a key learning curve, as it differs significantly from my academic experience. I’ve worked on various aspects of trading, applying data science and machine learning across business cases in fixed income and equity. Seeing the tangible impact of my work has been incredibly rewarding.
Since joining UBS, I’ve gained valuable experience in prototyping models and data pipelines, and more importantly: by productionizing, we’re making them practical and impactful for stakeholders.
Simone: I’ve always focused on smart analytics for order flow monitoring and automation, and enhancing client journey satisfaction across asset classes (like equity, fixed income and foreign exchange). Our best execution framework has become a cornerstone of our global regulatory setup.
Laurent: My main contributions in recent years have been designing and implementing a common framework for mass data movements and transformations. I’m proud that it’s now the backbone of our platform projects.
Zhilan: Since joining, I’ve worked on a variety of exciting projects that deepened my understanding of our team’s products and allowed me to explore new technologies like cloud and AI. These experiences have broadened my skill set and given me hands-on exposure to innovative tools and approaches, which I’ve applied to real-world challenges. It’s been a great mix of learning, growth and collaboration.
How do you maintain a fast-paced, agile approach within your team and the firm?
Simone: Three things make our agility possible: Scrum methodology, a platform that supports fast prototyping with production data and close collaboration between business and tech. My dual role helps bridge these worlds.
Zhilan: Adopting the Scrum agile methodology allows us to maintain a fast-paced and flexible approach. By breaking tasks into smaller, manageable sprints, we deliver incremental value while staying adaptable to changing priorities. This iterative process ensures continuous progress and improvement. Combining structured processes with open collaboration creates an environment that mirrors the agility and dynamism of a startup – while delivering high-quality results in a large organization like UBS.
Simone: We’re fortunate to have a dedicated sandbox for the Trading Analytics team, maintained by the Data Analytics Pod. Because it runs in the production environment, we can access real-world execution, client and market data across asset classes. At the same time, we’re not bound by the standard SDLC process, so our data scientists can rapidly prototype, iterate based on feedback, and experiment with new technologies – without the delays of regular release cycles.
Once a prototype is ready, moving it to production is seamless, thanks to the close collaboration between Trading Analytics team and the Data Analytics Pod from the very start.
Vito: This fast-paced environment really does feel like a startup. Our team of data scientists works closely with our tech teams, who keep our project infrastructure running smoothly. This collaboration is key. When building prototypes, we follow specific procedures to ensure that, when it’s time to productionize the model, we can refine and adapt it into a production-ready solution. This streamlined process drives efficient progress from start to finish.
Laurent: Working with experienced, open-minded colleagues on both IT and business sides allows us to challenge ideas and co-create solutions quickly. Agile tools and methodologies give us the structure to move fast. As Zhilan mentioned, working in sprints, breaking down complex tasks into smaller deliverables, and continuously reviewing progress and objectives helps us adapt quickly and maintain momentum.
In what ways are you encouraged to experiment with new ideas and approaches in your own playground?
Vito: I have the freedom to work with a highly flexible and well-equipped tech stack, which is fantastic. I can use any library I need, load and process data, implement models, test use cases, and ensure data quality. This freedom allows me to explore every idea – whether it’s testing a new model, processing data in a specific format, or optimizing runtime. The latter is especially important to ensure smooth handover to the TTA team without issues like memory overloads or inefficiencies.
Zhilan: Experimentation and innovation are actively encouraged through various initiatives and resources. For example, I was tasked with experimenting with a Microsoft Teams bot, which gave me hands-on experience and insight into its capabilities.
Vito: And there are also cross-team and global initiatives we can take part in…
Zhilan: Yes! Participating in internal hackathons is another great example. After developing products during these competitions, I was encouraged to share and showcase the solutions. This not only validated the work but also inspired further innovation. These opportunities encourage a spirit of discovery, where we can test ideas, learn from failures, and refine our approaches – all while being supported by the tools, platforms, and encouragement needed to succeed.