Andrea Kalenda

Master's Student in Supercomputing

Education

Master's program in Supercomputing

Brno University of Technology, Faculty of Information Technology

Brno, Czech Republic · 2025 - Present
Exchange Student

Seoul National University of Science and Technology

Seoul, South Korea · September 2024 - December 2024
  • Freemovers program, studied subjects based on machine learning
Bachelor's in Information Technology

Brno University of Technology, Faculty of Information Technology

Brno, Czech Republic · 2021 - 2025
  • Thesis: Detection of a Digital Addiction from Brain Signals

Technical Skills

Programming Languages
PythonC++CGoTypeScript
DevOps
CircleCIGitHub ActionsJenkinsTerraform
Tools
GitCursorJetBrains suiteLinuxBashCMakeMakeAWS
Libraries & Frameworks
PyTorchSklearnPandasNumpyOpenCVReactNextJSPyTestgtest
Languages
Czech (native)English (Fluent)Spanish (Basics)Korean (Basics)
Databases
PostgreSQLMongoDBRedisSQLiteSQLClickhouseElasticsearch

Work Experience

Associate Software Test Engineer

Solarwinds

Brno, Czech Republic · October 2022 - September 2024
  • Development and maintaining internal automation tool for testing written in Python utilizing Selenium and Pytest with features such as priority tests and management of timed runs outside of connected CircleCI
  • Cooperation with other teams on new features and accommodation of their testing approaches in our tool
  • Cooperation with DevOps on minimizing test run time, resources and general CI/CD pipeline management
  • Maintaining a Slack bot for engaging with the testing tool

Research Experience

Blockchain Research Assistant

Security@FIT, Brno University of Technology

September 2022 - July 2023

  • Investigation of a DAG chain based Kaspa cryptocurrency and whether it is susceptible to a set of specialized blockchain attacks
  • Exploration of Kaspa's unique consensus protocol

Projects

Custom CI/CD Pipeline Manager

A simple CI/CD pipeline management system with a web dashboard and Python backend, deployable via Docker Compose. Features include real-time pipeline monitoring, background job execution, live log streaming, and an auto-refresh dashboard. Pipelines are defined as JSON with configurable steps, variables, and timeouts. The frontend is built with TypeScript and React, offering a lightweight local alternative to hosted CI/CD tooling.

PythonTypeScriptReactCI/CDDocker
EEG-based Detection of Internet Addiction

A deep learning research project focused on classifying EEG recordings for signals associated with internet addiction, implemented in Python and PyTorch. The main model combined Transformer blocks with convolutional layers and residual connections, while several alternatives including LSTMs and raw CNN variants were evaluated to compare representation quality, stability, and overall classification performance. Bachelor thesis project.

PythonPyTorchDeep LearningEEG
Storehouse

A proof-of-concept storage and inventory management system built with Go microservices communicating via gRPC, orchestrated on Kubernetes. Includes an Inventory Service (product stock levels), Order Service (customer orders), with Docker containerization and automatic database schema initialization with sample data. Designed as an educational exercise in distributed systems, gRPC inter-service communication, and resilient architecture. Includes a testing client and PowerShell scripts for proto generation, image build, and K8s deployment.

GoKubernetesgRPCMicroservices
Turing Machine

A nondeterministic Turing machine simulator in Prolog using BFS to avoid infinite loops that would occur with DFS. States are dynamic predicates and the tape is a list of symbols; the model validates input, applies matching rules, and prints the output. Includes input validation (tape symbols, rule format, start/final state presence), exit code mapping for different failure modes, and example runs demonstrating empty-tape handling and loop avoidance. Built with SWI-Prolog and Make.

PrologAutomata Theory
Digital Piano Tone Synthesis

Advanced DSP for analyzing and synthesizing piano tones in Python. Covers MIDI notes 24–108 with autocorrelation for lower tones and DFT-based spectral analysis for higher ones, plus DTFT refinement for cent-level accuracy. Extracts magnitude and phase for the first five harmonics, reconstructs tones via inverse DTFT, and generates complete musical pieces from note sequences at 48 kHz and 8 kHz. Uses SciPy, NumPy, Matplotlib, and SoundFile for processing and visualization.

PythonDSPSignal Processing
Simulation Model of SpaceX's Mars Logistics

A discrete-event simulation (C++ with SIMLIB) of the full Mars logistics chain: rocket assembly, FAA authorization, orbital refueling, and interplanetary transit. Models Starship lifecycle, Super Heavy Boosters, launch pads, tankers, and the 780-day Mars transit window. Configurable parameters for fleet size, payload, and infrastructure. Key findings: delivering 10,000 tons requires ~2.4 years minimum; 3 launch pads and 3 tankers are optimal; Mars window frequency is the main bottleneck. Includes documentation and experimental validation.

C++SimulationSimlib
View all 9 projects

Competitions

Cyber Heroes Academy

Binary Confidence, 2025

Joined a team of 5 as a SOC analyst, actively trying to prevent multiple types of real-time cyber attacks coming from the red team.

Extracurricular Activities

Archery club member in a local club · 2019 - 2023
Member in VUT Erasmus Student Network · 2026 - present
Built with Astro, React & shadcn/ui