Keynotes | September 30
Paulo Marques
Paulo Marques is an entrepreneur, engineer, and investor. Paulo co-founded Feedzai, one of the largest AI/ML fraud prevention companies in the world, where he served as Chief Technology Officer (CTO) for 13 years, leading its product and technology strategy, and where he still serves on the board of directors. More recently, he’s one of the founding partners of TUMO Portugal, a revolutionary school for kids aged 12 to 18, covering the interconnection between creativity and technology. He is currently a Scientific Director of the CMU|Portugal program and an early-stage investor in several technology companies. Before, Paulo was co-Director of the CMU|Portugal program in Software Engineering, having a dual appointment as professor at Carnegie Mellon and at the University of Coimbra. Paulo holds a PhD in Distributed Systems, has authored over 40 peer-reviewed papers as well as a book. On his free time Paulo likes to fly planes.
Evgenia Smirni
Evgenia Smirni is the Sidney P. Chockley Professor and Chair of Computer Science at William and Mary, Williamsburg, VA, USA. Her research interests include queuing networks, stochastic modeling, data centers and cloud computing, applied machine learning, dependability and reliability, workload characterization, and models for performance prediction of distributed systems and applications. She has served as the Program Co-Chair of QEST’05, ACM Sigmetrics/Performance’06, HotMetrics’10, ICPE’17, DSN’17, HPDC’19, and SRDS’19, and the cloud track Co-Chair of ICDCS’21. She served as the General Co-Chair of QEST’10, NSMC’10, ICPE’25, and as general chair of SIGMETRICS’23. She is an IEEE Fellow, an ACM Distinguished Scientist, and an elected member to the IFIP W.G.s 7.3 and 10.4.
Securing the Machine Learning Components of Autonomous Vehicles: Risk Assessment and Mitigation
Autonomous vehicles (AVs) are one of the most complex software-intensive Cyber-Physical Systems (CPS). In addition to the basic car machinery, they are equipped with driving assistance mechanisms that use smart sensors and machine learning (ML) algorithms for perception of the environment, pathfinding, and navigation. Even though tremendous progress has been made in advancing safety and security of AVs, they are shown to be vulnerable to accidental and malicious faults that negatively affect their perception and control functionality and result in safety incidents. With increasing use of specialized hardware accelerators such as GPUs for running ML-based perception algorithms, AV control systems have become susceptible to transient faults (soft errors) that can result in erroneous ML inference and impact their decision making and control. In addition, safety assurance for AVs requires testing their end-to-end resilience against most salient safety-critical faults and attacks targeting their controller inputs, software/hardware stack, and output within realistic simulation environments to reduce cost of road testing and risk of harm to drivers and pedestrians. In this talk, I will outline a new methodology that can assess the ML components and AV control systems and illustrate ways to fortify them against accidental or malicious faults. This is joint work with Anne Schmedding, Yiyang Lu, Xugui Zhou, Homa Alemzadeh, and Lishan Yang.
Keynote | October 1
Jacopo Tagliabue
Jacopo Tagliabue is the co-founder and CTO of Bauplan. Educated in several acronyms across the globe (UNISR, SFI, MIT), Jacopo previously co-founded Tooso, an AI startup acquired by TSX:CVO in 2019. He led AI efforts at Coveo from scale-up to IPO and built Coveo Labs, a prolific R&D practice whose open libraries, models, and datasets have been downloaded millions of times.
Throughout his career, he has been fortunate to collaborate with remarkable folks in both industry and academia (e.g., Netflix, NVIDIA, Stanford Univ., UChicago, Univ. of Wisconsin-Madison), and contribute to diverse fields including Information Retrieval (RecSys, SIGIR), Data Science (KDD), Artificial Intelligence and NLP (ICML, NAACL), Data Management (SIGMOD, VLDB), and Computer Systems (Middleware). While building his company, he teaches ML Systems at NYU, which is notable (mostly) because it is the only job he ever had that his parents understand.
How Do We Sleep at Night? Building Reliable Distributed Systems at Startup Speed
Bauplan is a data lakehouse that leverages a purpose-built Function-as-a-Service runtime to perform data transformations, analytical queries, and import jobs. While users enjoy a simple, Docker-like experience directly in their terminal, a single Bauplan command may orchestrate operations across more than half a dozen cloud systems, ranging from endpoints and databases to caches, virtual machines, and object storage.
These components span a wide spectrum of reliability: from hyper-scaler stability to obscure open-source projects we wrap, to fully custom libraries. Every day, tens of thousands of Bauplan jobs run successfully in production for large enterprises and fast-growing startups, raising the question: how do we sleep at night?
In this talk, we share our lessons from building distributed systems at startup speed and production scale. We discuss how to reuse open-source software without blindly trusting it, implement end-to-end checks, and invest in foundational technologies. Finally, we show how we embedded simulation and formal verification into our product, illustrating how even small companies can collaborate with academia to accelerate their software lifecycle.