Professional Experience

• Research Associate @ IBM Research - Zurich (2025) [9mo]
  • Systems/LLM arch. co-design resesarch in position-independent caching, achieving:
  • • "Best Research Paper Honorable Mention" award at top-tier ML Systems conference (MLSys 2026)
  • • unique open-source vLLM-native implementation (sub-100 LOC footprint)
  •   due to caching abstractions better-aligned with vLLM abstractions
  • • improvements in KV cache hit rates (causing 10x TTFT reduction)
  • • long-context cold-start prefill speedups (~2.5x on H100)
  • • detailed evaluations / output quality study
  • • Explored state combination methods for SSM hybrid LLM architectures
  •   (SSM, Liquid, LightningAttn.)
  • • Supervised by early and major vLLM contributor
  • └> Resulted in public release as IBM's official vLLM public fork
  • └> + Arguably the fastest end-to-end implementation of LLM inference
          with this feature among major labs
  • └> + Subsequent work by international teams within IBM Research
• Project & Design Lead @ Hololearn TU Delft research group (2022) [5mo]
  • Based on EU-level research grant
  • Assembled and led a 5-person team to develop and
    design hologram-based virtual teaching system
  • Involved full-stack development including 3D+LiDAR data processing
  • Attracted collaboration from Technologico de Monterrey
  • └> led to further development + CEUR Workshop publication (6 citations)
• Founding Data Scientist @ CorTexter (2020-2021) [1y]
  • Developed+designed core ML pipeline for advanced
    NLP-based (neural + linguistic) job vacancy analysis
  • Trained emoji-prediction language model
  • Built multilingual neural job title search engine
  • └> pipeline still in use after leaving
• Startup Analyst @ Plug&Play Tech Center (2020-2021) [9mo]
  • Major California-based VC fund
  • Analyzed hundreds of startup pitch decks
  • └> Praised for story-driven startup review by top analysts

Research Highlights

• MLSys Paper (2026):
  • "Using Span Queries to Optimize for Cache and Attention Locality"
  • received "Best Research Paper Honorable Mention" award
  • Introduced span queries language as expression trees for inference calls that use
  • pos. indep. caching, with commutativity constraints, generalizing pos. indep. caching
  • LLM serving beyond RAG to inference-time scaling
• Msc. Thesis (2024):
  "Towards faster sequence-to-sequence models for basecalling"
  Introduced novel DL architectures for DNA basecalling
  (up to 85% throughput improvement over best Oxford Nanopore model)
  Introduction of Chain-of-thought technique to basecalling (improving accuracy-ceiling of pareto optimal models)
  └> led to partnership with Tenstorrent (AI accelerator company)
• IJEPES Paper (2024) (65 citations):
  "PowerFlowNet: Power flow approximation using message passing Graph Neural Networks"
  International Journal of Electrical Power & Energy Systems
  GNN-based power network prediction model
  achieving 48x speedup over non Deep-Learning-based method (Newton-Raphson)
  └> led to further work by co-authors
• Arxiv Preprint (2023):
  "Lights out: training RL agents robust to temporary blindness"
  Pioneered RL loss allowing an agent to navigate
  through temporary blindness (open loop control)
  └> led to a successful research grant

Achievements & Projects

• - Ready to Startup jury winner for space transport platform
    (validated by NASA contractor Axiom Space)
• - 1st place Yes!Delft Students hackathon for coral reef startup
• - 3rd place Best Delft ideation contest
• - Selected Nova Network member (top ~3% of applicants)
• - Host of UvA/Slim Radio Monday Morning Podcast
    interviewing high-achieving students