Tomáš Moždřeň
IT Systems Administrator & Home Lab Enthusiast
Czech Republic 🇨🇿
About
IT Systems Administrator currently working at ČEZ Distribuce, maintaining enterprise systems running the Apache Big Data stack (Hadoop, Kafka, Ozone, etc.) on RHEL 9. Responsible for system maintenance, infrastructure automation with Ansible, user support, and troubleshooting in a corporate environment.
Outside of work, I maintain a Proxmox-based home lab with ZFS storage, GPU passthrough, and containerized services. I build efficient, maintainable systems—from enterprise deployments to custom C utilities. Minimal overhead, maximum reliability.
Experience
-
Analytics and Data Platforms Specialist — ČEZ Distribuce
2025 - Present
Maintaining Apache Big Data stack (Hadoop, Kafka, Ozone) on RHEL 9, Ansible automation, user support.
Technical Skills
Virtualization & Infrastructure
- Proxmox VE
- ZFS (pools, ARC tuning, mirrored setups)
- Docker & LXC containers
- KVM/QEMU
- Ansible
Linux Administration
- RHEL 9 / CentOS
- Debian / LMDE / Ubuntu Server
- systemd & crontab automation
- Samba file sharing
- Networking (firewall, DNS, VLANs)
Big Data & Enterprise
- Apache Hadoop
- Apache Kafka
- Apache Ozone
Self-Hosted Services
- Nextcloud
- Jellyfin / Plex
- Wiki.js
- Navidrome
- Home Assistant
- Ollama (local LLMs)
Programming & Scripting
- Bash / Shell scripting
- Python
- C (system utilities)
- SQL
- Git
Languages
- Czech (native)
- English (fluent)
Projects
- BeanMon — Lightweight X11 system monitor (CPU, RAM, disk) written in C
- tin — Minimal text editor in C
- tinterm — Fast terminal emulator using VTE
- tinshell — Custom shell in C
- lineprint — Faster cat alternative for terminal output
- Work-Term — Pomodoro timer in Python/tkinter
More on GitHub
Infrastructure Projects
Home Lab — "starhaven"
Infrastructure Overview
7 VMs + 4 LXC containers running on Proxmox VE 9.1.4 (kernel 6.17.4-2-pve)
Hardware
- Server: Dell Precision 3630 Tower
- CPU: Intel Core i5-8500 @ 3.00GHz (6 cores)
- RAM: 64GB DDR4 2666MHz (4x16GB)
- GPU: NVIDIA GTX 1050 Ti (passthrough to Jellyfin)
Storage — 9TB Total
- NVMe: Crucial P1 1TB — Proxmox OS, LVM-thin (794GB available)
- ZFS Pool: WD 8TB @ 7200rpm (vm-8tb-hdd0) — VMs (4TB+ used)
- Backup HDD: WD Blue 1TB @ 7200rpm
- ZFS ARC: 1-24GB tuned (min 1GB, max 24GB)
Services
| VM/Container | Purpose | Resources |
|---|---|---|
| media-server | Jellyfin + GPU transcoding | 4 cores, 4GB, 250GB + 2TB |
| navidrome-bean | Music streaming | 2 cores, 2GB, 10GB + 1.95TB |
| dev.bean | Development environment | 2 cores, 4GB, 26GB |
| dev.remote | Guacamole remote access | 2 cores, 4GB, 26GB |
| prometheus | Metrics collection (LXC) | 1 core, 256MB, 8GB |
| grafana | Monitoring dashboards (LXC) | 1 core, 256MB, 8GB |
| proxexport | Proxmox metrics exporter (LXC) | 1 core, 256MB, 8GB |
| uptimekuma | Uptime monitoring (LXC) | 1 core, 512MB, 8GB |
Monitoring & Observability
- Uptime Kuma: Service uptime monitoring
- Grafana: Real-time infrastructure dashboards
- Prometheus: Metrics collection and alerting
- Proxmox Exporter: Proxmox VE metrics integration
Planned Upgrades
- CPU: Upgrade to Intel Core i7-8700 (6C/12T)
- RAM: Expand to 128GB DDR4
- Services: Nextcloud VM, [Wiki.js](http://wiki.js/), and more...
- Download my custom grafana dashboard: Grafana Dashboard (Proxmox Starhaven Homelab)
Smart Home — Raspberry Pi 4
Hardware
- Device: Raspberry Pi 4 (BCM2711, 4GB RAM)
- Optimizations: CPU scaling, ZRAM, resource limits
- Uptime: 99.9% reliability
Stack
- Home Assistant
- Node-RED
- Zigbee2MQTT
- Docker
- Apache
- MQTT
Integrations
- HACS: Local WebRTC (secure local video), Local Tuya (cloud-free controls), MQTT clients
- Devices: 20+ smart devices managed via Zigbee
- Automation: Custom Node-RED workflows for family smart home