Before we begin: My research papers are generally very formal and full of math, but this site is designed for a broader audience. Pictograms are used here as light visual cues to make the content easier and a little more fun to follow.

👋 Welcome!

I am Filip Marković, a lecturer (equivalent to assistant professor) in Computer Science at the University of Southampton, UK 🇬🇧. Previously, I was a postdoctoral researcher at the Max Planck Institute for Software Systems (MPI-SWS) in Kaiserslautern, Germany 🇩🇪, working with Björn Brandenburg. Before that, I pursued my PhD and postdoctoral research at Mälardalen University in Västerås, Sweden 🇸🇪. I am originally from Podgorica, Montenegro 🇲🇪.

🔎 Research Vision

My research focuses on uncertainty-aware analysis and scheduling of real-time and cyber-physical systems. As these systems grow in complexity, so does their inherent timing uncertainty, which has already contributed to a series of safety incidents ⚠️ in recent years. This trend is further amplified by AI-powered components, heterogeneous hardware, and adaptive software layers. At the same time, traditional methods for assessing timing-related safety properties and guiding principled resource provisioning are becoming increasingly inadequate.

In practice, this challenge is critical across many domains, including automotive 🚗, avionics ✈️, medical devices 🏥, and aerospace and space exploration 🚀. For example, in autonomous vehicles and drones, detecting an obstacle is important, but ensuring the system reacts on time ⏳ to avoid a collision ⚠️ is essential. As modern functionalities (such as computer vision, machine learning, and artificial intelligence) demand ever more computational resources, it becomes increasingly challenging to guarantee correct timing behavior 📟 + ⏳ → ☑️.

⏱️ From Scheduling Theory to Space Exploration 🚀

In my research, I use statistical inference 📈, probability theory 🎲, machine learning 💻➕🎓, randomized algorithms 🎲➕🧮, mechanized formal verification 🔢➕⚙️, and related methods, to analyze the uncertainty in the timing behavior of complex cyber-physical systems and to develop scheduling algorithms that are explicitly uncertainty-informed.

My research has resolved several long-standing open problems in uncertainty-aware analysis and scheduling of cyber-physical systems, resulting in six papers at RTSS (a flagship conference, ranked A* in the ICORE system), and four at other A-ranked venues, including two Outstanding Paper Awards (at RTSS 2025 and ECRTS 2021).

These developments also have a broader practical impact. For example, I collaborate with NASA Ames Research Center on our timing-inference tool LiME, which we are actively developing for prospective integration into their safety procedures and timing validation pipeline. I also contribute to the NASA’s suborbital mission ADAPT, where we develop uncertainty-aware scheduling algorithms for the timely observation of transient astronomical phenomena, including gamma-ray bursts, novae, and supernovae.

You can see me below at NASA Ames, where I was invited for a research visit in May 2025. This occasion symbolically united my scientific work with my childhood passion for astronomy (as my grin makes it clear 😁). Alongside this, I share a few more figures from my recent papers at the intersection of scheduling theory and astronomy. I leave them intentionally unexplained here, in the hope that they spark your curiosity to explore my publications and perhaps ignite a discussion if we meet in person.

Finally, I wish to acknowledge with gratitude my colleague Marion Sudvarg for inviting me to contribute to the ADAPT mission, which has already led to one RTSS’25 publication (to be presented soon) and two at ICRC’25. I am equally thankful to Irfan Šljivo for making possible my research visit to NASA Ames, where our collaboration on LiME began. Both efforts are deeply rewarding and represent just the start of an exciting research journey.