resume.tex

\documentclass[10pt]{article}
\usepackage{array, xcolor, lipsum, bibentry}
\usepackage{scrextend}
\usepackage[margin=1.0cm]{geometry}
\usepackage{fancyhdr}
%\usepackage[czech]{babel}
\usepackage{scrextend}
\usepackage{hyperref}

\geometry{
	a4paper,
	bottom=10mm,
	top=10mm,
}

\begin{filecontents}{publication.bib}
	@inproceedings{letal2015finding,
		title={Finding New Malicious Domains Using Variational Bayes on Large-Scale Computer Network Data},
		author={Letal, Vojtech and Pevny, Tomas and Smidl, Vaclav and Somol, Petr},
		booktitle={Advances in Approximate Bayesian Inference, NIPS 2015 Workshop},
		year={2015}
	}
\end{filecontents}

\begin{document}
\pagenumbering{gobble}% Remove page numbers (and reset to 1)
\clearpage
\thispagestyle{empty}
%\maketitle
{\center\bfseries\huge Vojtech Letal}
\hfill
{\bf Email:} letal.vojtech@gmail.com
{\bf Tel:} +41799314964
{\bf Github:} \href{http://github.com/letalvoj}{letalvoj}

\section*{About me}
I'm a passionate ML engineer with experience building and deploying models. I'm seeking new opportunities to help people make the most of LLMs in daily life.

\section*{Professional Experience}
\subsubsection*{\bf Software Engineer, Google (Zurich, Switzerland) \hfill 09/2022--present}
    \begin{itemize}
        \setlength\itemsep{-0.25em}
        \item Main contributor to the launch of a new Gemini feature that allows users to select and modify response parts (achieving a 90\% response acceptance rate).
        \item Curated high-quality datasets for fine-tuning and reward modeling using annotation and model distillation.
        \item Trained and evaluated handwriting recognition transformer models for handwritten math recognition (20\% project).
    \end{itemize}

\subsubsection*{\bf Staff Scientist, AI - Avast s.r.o. (Prague, Czechia) \hfill 04/2021--08/2022}
	 \begin{itemize}
		\setlength\itemsep{-0.25em}
		\item Personally recognized by the CEO Ondrej Vlcek as one of a small group of employees with high growth potential.
		\item Responsible for internal productization of a deep learning platform based on Mill.jl co-developed by Avast and CTU.
		\item Developed and maintained a DL pipeline that classified malware samples, processing \textbf{all} user-reported malware samples in near-real time.
	 \end{itemize}

\subsubsection*{\bf Senior ML Engineer \& Tech Lead - Blindspot.ai (Prague, Czechia) \hfill 09/2015--04/2021}
	 \begin{itemize}
		\setlength\itemsep{-0.25em}
		\item Leading a team of developers working on several analytical and ML tasks for a US-based cybersec startup.
		\item Responsible for developing a real-time event-based anomaly detection platform.
		\item Contributed to the popularization of regression modeling for live election counts in Czechia.
	 \end{itemize}

\subsubsection*{\bf DS12 Resident - DataScience Inc. (Culver City, CA, USA)\hfill summer 2016}
	 \begin{itemize}
		\setlength\itemsep{-0.25em}
		\item 1 of 9 residents selected to participate in a 12-week elite, intensive residency program with 2.5\% acceptance rate.
        \item Implemented recommendation system, churn prediction model, and scalable ETL pipeline costing 36x less per run.
	 \end{itemize}

\subsubsection*{R\&D - Cisco Systems (Prague, Czechia)\hfill 06/2014--08/2015}
	 \begin{itemize}
		\setlength\itemsep{-0.25em}
		\item Implemented novel Bayesian inference algorithm to estimate probability of maliciousness of domains.
		\item Extensively used big data technologies such as Twitter Scalding and Apache Spark.
	 \end{itemize}

\section*{Education}
\subsubsection*{\bf Master’s degree in Artificial Intelligence \hfill 09/2012--06/2015}
\vspace{-1.5mm}
Czech Technical University in Prague - Open Informatics \textit{(First-class honors)}

\section*{Knowledge}
\begin{labeling}{Programming}
	\setlength\itemsep{-0.15em}
    \item [\textbf{Languages}] Czech (Native), English (Fluent)
    \item [\textbf{Programming}] Python, C/C++, Java/Kotlin, Scala, Bash, Git, SQL, Angular
    \item [\textbf{Frameworks}] Tensorflow, Scikit-Learn, Pandas, Jupyter
    \item [\textbf{Skills}] LLMs, Prompt Engineering, Parameter-Efficient Methods, Machine Learning
\end{labeling}

\section*{Patents}
Discovering yet unknown malicious domains using relational data. {\it US Patent: 10,320,823}
\\
Modifying Source Code Using Generative Models. {\it US Patent: Pending}

\bibliographystyle{plain}
\nobibliography{publication}

\section*{Publications}
\bibentry{letal2015finding}.

\vfill

\end{document}