Curriculum Vitæ (short version)

Curiosity-driven Senior Software & ML Engineer with a strong background in Theoretical Computer Science. Proven track record of co-authoring production-grade AI tools (H2O Sonar) and architecting core algorithms for globally recognized open-source AutoML frameworks (H2O-3). Expert in Machine Learning Explainability (XAI), advanced statistical modeling (GLM/Tweedie), and building advanced evaluation systems for Generative AI (LLMs, RAG, and Agentic architectures).

• Languages: Python, Java, R, Clojure, Emacs Lisp, Common Lisp, C/C++
• Machine Learning & AI: AutoML, Generative AI, LLM Evaluation, RAG & Agentic Systems, Explainable AI (SHAP), Statistical Modeling (GLM, Tweedie), Conformal Prediction, Probabilistic Programming (Stan, PyMC)
• Tools & Infrastructure: Linux, Git, Docker, K8s, MLOps, Data Warehousing, Emacs/Org-mode

2020 – now

Senior Software & ML Engineer @ H2O.ai

• Generative AI & Evaluation: Co-authored H2O Sonar (GenAI Evaluators) and architected core ML components for H2O Eval Studio, establishing rigorous frameworks to evaluate Large Language Models (LLMs), RAG architectures, and agentic systems.
• Open-Source AutoML: Core developer of open-source AutoML and Stacked Ensembles within H2O-3 across Java, Python, and R, optimizing both algorithm performance and system scalability.
• Explainable AI (XAI): Researched and implemented core mathematical methods for machine learning interpretability; created the explain module and integrated Marginal/Interventional SHAP algorithms (TreeSHAP, DeepSHAP, Generalized DeepSHAP) into the open-source H2O-3 engine.
• Advanced ML Modeling: Implemented advanced interpretable algorithms, including AdmissibleML and a complex Tweedie family estimator for GLM with simultaneous estimation of dispersion and variance power (PR 1, PR 2).
• Ecosystem Maintenance: Acted as the official maintainer of the H2O-3 R package, ensuring ecosystem stability and package governance.

2018 – 2020

Freelance Data Scientist & Team Lead @ Datamole & Consulting

• Team Leadership: Promoted to Data Science Team Lead, managing engineering priorities and client delivery.
• Engineering & MLOps: Architected custom optimization routines, contributed to an internal MLOps system, and co-designed enterprise data warehouses.
• Academic Teaching: Taught Distributed Data Mining at the Czech Technical University (FIT CTU).
• Consulting: Provided independent advisory services in information theory, statistics, and graph theory.

2017 – 2018

Data Scientist @ Vendavo

• Price Optimization: Engineered price optimization and economic models using classical statistical methods in R.
• Collaboration: Partnered with an interdisciplinary PhD-heavy team bridging advanced Econometrics and Artificial Intelligence.

Jul 2016 – Sep 2016

Research Assistant @ FIT (VýLeT)

• Research Focus: Researched scalable machine learning of predictive ensembles using the FAKE GAME approach on the H2O platform, which directly laid the foundation for my award-winning Master's thesis and publication in Machine Learning.

2014 – 2017

Master's in Knowledge Engineering, Faculty of Information Technology, Czech Technical University (FIT CTU)

• Thesis: Scalability of Predictive Modeling Algorithms – Awarded the Dean's Award (Winter 2016/2017)

2011 – 2014

Bachelor's in Theoretical Computer Science, Faculty of Information Technology, Czech Technical University (FIT CTU)

• Thesis: Comparison of different neuroevolution algorithms on several problem domains

2010 – 2011

Mathematical Informatics, Faculty of Nuclear Sciences and Physical Engineering, CTU

• Transferred to FIT for closer alignment with Computer Science interests.

Full registry available at ORCID and Google Scholar.

• Nykodym, Tomas and Kraljevic, Tom and Wang, Amy and Wong, Wendy and Fryda, Tomas (2025). Generalized linear modeling with h2o, Published by H2O.ai Inc. Available here.
• Kordík, Pavel, Černý, Jan, Frýda, Tomáš. (2018). Discovering predictive ensembles for transfer learning and meta-learning. Machine Learning. 107. doi: 10.1007/s10994-017-5682-0.
• Kordík, Pavel, Frýda, Tomáš (2018). On Scalability of Predictive Ensembles and Tradeoff Between Their Training Time and Accuracy. In: Shakhovska, N., Stepashko, V. (eds) Advances in Intelligent Systems and Computing II. CSIT 2017. Advances in Intelligent Systems and Computing, vol 689. Springer, Cham. doi: 10.1007/978-3-319-70581-1_18.
• Kordík, Pavel, Frýda, Tomáš, Šnorek, Miroslav, Čepek, Miroslav. Scalability of predictive ensembles, 2017 12th International Scientific and Technical Conference on Computer Sciences and Information Technologies (CSIT), Lviv, Ukraine, 2017, pp. 555-560, doi: 10.1109/STC-CSIT.2017.8098848.
• Frýda, Tomáš (2017). Scalability of Predictive Modeling Algorithms, České vysoké učení technické v Praze. Available here.
• Frýda, Tomáš (2014). Comparison of Different Neuroevolution Algorithms on Several Problem Domains, České vysoké učení technické v Praze. Available here.