Martijn de Vos
Scholar

Martijn de Vos

Google Scholar ID: w2gzvWEAAAAJ
Postdoctoral Researcher, EPFL
Distributed systemsdecentralized systemsdecentralized learning
Citations & Impact
All-time
Citations
971
 
H-index
9
 
i10-index
9
 
Publications
20
 
Co-authors
16
list available
Resume (English only)
Academic Achievements
  • - Publications:
  • - 2025 Middleware’25: 'Leveraging Approximate Caching for Faster Retrieval-Augmented Generation'
  • - 2025 ICML’25: 'Robust ML Auditing using Prior Knowledge'
  • - 2025 WWW’25: 'Boosting Asynchronous Decentralized Learning with Model Fragmentation'
  • - Organizing activities:
  • - 2025 DICG'25 International Workshop (co-located with ICDCS conference)
  • - Services:
  • - 2025: Reviewer/PC member for multiple journals and conferences
  • - 2024: PC member for multiple conferences
  • - 2023: Organizer of DICG conference
  • - 2022: Organizer of DICG conference
  • - 2021: Published in Computer Networks journal
  • - 2020: Published in NDSS conference
  • - 2018: Published in IECON conference
Research Experience
  • - Postdoctoral researcher in the Scalable Computing Systems (SACS) Laboratory at EPFL University, focusing on building efficient and scalable machine learning systems, particularly within data centers (e.g., LLM inference, fine-tuning, and training), as well as ML over the Internet (federated and decentralized learning).
Education
  • - PhD: Completed in 2021, thesis titled 'Decentralization and Disintermediation in Blockchain-based Marketplaces', supervised by Dick Epema and Johan Pouwelse.
  • - Master's: Completed in 2016 at Delft University of Technology, thesis titled 'Identifying and Managing Technical Debt in Complex Distributed Systems', improved various aspects of the long-running academic software, Tribler.
Background
  • Research interests include machine learning systems, multi-agent LLM systems, distributed and decentralized systems, and distributed ledger technology. Currently a postdoctoral researcher in the Scalable Computing Systems (SACS) Laboratory at EPFL University, focusing on building efficient and scalable machine learning systems.
Miscellany
  • Besides professional activities, he enjoys traveling, snowboarding, bouldering, running, and reverse engineering software. One of his projects involves the emulation of legacy Apple devices, and the software can be found on his website. He also has a small company named CodeUp, primarily developing mobile applications (iOS/Android) and providing web/email hosting services.