Terezia Slaninakova, Ph.D.

Terezia Slaninakova, Ph.D.

Researcher at Masaryk University

About me

Hi! I am Terezia Slaninakova, Ph.D., a researcher at Masaryk University, Brno, Czech Republic. I build useful research production-ready applications at the intersection of machine learning, similarity searching, and bioinformatics. I am currently working at cerit-sc department on scientific discovery services within the eosc-cz project. If you’re working on interesting problems in vector search, RAG systems, scientific ML, or related areas, I’d love to connect!

My Ph.D. was dedicated to creating and improving a ML-based index for similarity search in high-dimensional data under the GF23-07040K project co-shared with CAU in Kiel, Germany. The theoretical work ultimately lead to creating a similarity search system for finding proteins with similar structure in 500M+ proteins and protein domains, AlphaFind - https://alphafind.ics.muni.cz, used by researchers worldwide. While on Ph.D., I attended three internships - in Daisy lab under prof. Jensen in Aalborg, Denmark, in Database Sysems and Data Mining group under prof. Kröger in CAU, Kiel, Germany, and, most recently, in Greener group at MRC-LMB in Cambridge, UK. I also colaborated with David Sehnal from CEITEC, MU on MolViewStories.

For my M.S.c, I studied Artificial intelligence and Data processing, while working on projects at CERIT-SC, where I created ML pipelines for life tissue and remote sensing data. For my thesis, I created a prototype of an ML-based index for metric spaces.

I got my B.S.c-equivalent in Computer Science in Brno University of Technology in 2018, in my final thesis I focused on generative AI applied to sound in Speech@FIT lab. While studying I also worked at Y Soft and Adacta on creating automated quality assurance pipelines.

Core Expertise

  • Similarity Search & Retrieval: Vector databases, ML-based indexing, hybrid search architectures, semantic search systems

  • ML Systems: PyTorch, embedding models, data processing pipelines, evaluation frameworks

  • Infrastructure: Kubernetes, Docker, batch processing, production deployment

  • Domains: High-dimensional data, search engines, scientific data, protein structure analysis, molecular dynamics