C-01
SOFTWARE ENGINEERING
Web applications, APIs, microservices, CLI tools and embedded systems. Production code — with tests, observability and documentation.
STATUS · OPERATIONAL · 24/7
Proxmox. Ceph. Kubernetes.
BGP. ZFS. eBPF.
We design, build and run the infrastructure your business stands on — Proxmox, Ceph, Kubernetes, networks, and our own code where it's needed. With documentation a human can read, and without slides about digital transformation. It works — and we make sure it stays that way.
- VirtualizationProxmox · Ceph · ZFS
- ContainersKubernetes
- NetworksBGP · WireGuard
- AutomationAnsible · Terraform
SLA
99.98%
Response
< 24h
Coverage
24/7
Region
PL·EU
Experience
10+ yrs
00
LIVE TELEMETRY
T+00:00:00 SYS nominal — all clusters reporting T+00:00:00 K8S 12/12 nodes ready · control-plane healthy · etcd quorum 3/3 T+00:00:00 VIRT hypervisor cluster 3/3 online · storage healthy · uptime 287d T+00:00:00 NET bgp sessions established · tunnels up · loss 0.00% T+00:00:00 OBS prometheus scraping · 0 critical alerts · p99 8.4ms T+00:00:00 BKP last successful run 18m ago · retention 30d · offsite ok T+00:00:00 SLA 99.98% rolling 30d · within budget
01
CAPABILITIES
Full lifecycle —
from design to operations.
No middle layer, no "project managers who don't know what they're talking about". Engineers work directly with your team.
C-02
KUBERNETES OPERATIONS
We design production clusters and operate them long-term. Self-hosted, managed (EKS / GKE / AKS) and hybrid.
C-03
VIRTUALIZATION & PRIVATE CLOUD
Hyperconverged virtualization environments as a cost-effective, full-featured alternative to commercial platforms. Storage, networking, backup, replication — without compromise.
C-04
NETWORK ENGINEERING
Design and deployment of LAN / WAN, VPN, segmentation, firewalls and VLANs. Where packets matter.
C-05
DEVOPS & AUTOMATION
CI/CD, IaC, GitOps. We replace manual work with repeatable, auditable pipelines.
C-06
LINUX SYSTEMS
Bare-metal, VPS, dedicated servers. Hardening, patching, monitoring, capacity planning.
C-07
OBSERVABILITY
A stack that tells the truth — before the customer calls. SLO, error budgets, on-call rotation, alerts without false positives.
C-08
CLOUD & HYBRID
AWS, GCP, Azure, Hetzner, OVH. We pick the cloud for the problem, not for the marketing. Often, owning the iron is cheapest.
C-09
PERFORMANCE ENGINEERING
Profiling, hot-path optimization, zero-copy I/O, eBPF, io_uring. Where milliseconds cost money.
02
TECHNOLOGY STACK
Right tool for the problem,
not for the trend.
LANGUAGES
Go · Rust · Python · TypeScript · C · C++ · Bash · Lua · Zig
ORCHESTRATION
Kubernetes · Docker · Podman · Helm · ArgoCD · Flux · Kustomize · Cilium · Istio
VIRTUALIZATION
Proxmox VE · KVM/QEMU · LXC · Ceph · ZFS
OPERATING SYSTEMS
Debian · Ubuntu · RHEL · Rocky · AlmaLinux · NixOS · FreeBSD
NETWORKING
MikroTik RouterOS · OPNsense · pfSense · FortiGate · WireGuard · IPsec · BGP · OSPF · VLAN · nftables
AUTOMATION / CI-CD
Terraform · OpenTofu · Ansible · Pulumi · GitLab CI · GitHub Actions · Jenkins · Tekton
OBSERVABILITY
Prometheus · Grafana · Loki · Tempo · VictoriaMetrics · OpenTelemetry · ELK · ClickHouse
DATA / MESSAGING
PostgreSQL · MariaDB · Redis · MongoDB · Kafka · RabbitMQ · NATS · MinIO
CLOUD
AWS · GCP · Azure · Hetzner · OVHcloud · Cloudflare
03
ENGAGEMENT PROCESS
Four phases.
Clear scopes. Written commitments.
-
PHASE 01
RECONNAISSANCE
Technical audit, understanding of the problem and constraints. You receive a document: what we see, what we propose, what we won't touch without your approval.
-
PHASE 02
DESIGN & QUOTE
Architecture, schedule, scope, costs. Fixed-price or hourly. An honest forecast — even at the cost of a smaller contract.
-
PHASE 03
DEPLOYMENT
Short iterations. Every deployment under version control, every change auditable. Full technical documentation at the end.
-
PHASE 04
SUSTAINMENT
After deployment: monitoring, updates, incident response, evolution. SLA tailored to actual needs, not to a price list.
04
CAPABILITY MATRIX
Specifically what we do.
Tasks that are Tuesday work for us — not quarterly research projects.
M-001
Stand up an HA Kubernetes cluster on bare-metal in days, and operate it for years.
K8S
M-002
Migrate from VMware vSphere to a self-hosted hyperconverged platform with minimal downtime.
VIRT
M-003
Design a network covering office, colocation and remote workforce in a single coherent model.
NET
M-004
Introduce GitOps where today deployments happen via SSH and file copies.
DEVOPS
M-005
Rewrite a critical Python module to Go and cut server bills by 70%.
SW
M-006
Stand up a backup system resilient to ransomware (3-2-1, immutable, off-site).
BKP
M-007
Deploy monitoring that doesn't generate false alarms and doesn't wake people up for nothing.
OBS
M-008
Take on-call duty for a team that would finally like to sleep on weekends.
SRE
M-009
Design and deploy a WireGuard mesh connecting offices, clouds and people.
NET
M-010
Optimize the PostgreSQL queries no one wanted to touch.
DB
M-011
Automate end-to-end server provisioning (Terraform + Ansible + cloud-init).
IAC
M-012
Run a full security audit of infrastructure and network with a concrete remediation plan.
SEC
05
SECTORS
Who we work with.
A small company whose server room is a rack under the desk. A multinational SaaS with infrastructure in four regions. We fit.
S-01
SAAS & SOFTWARE
We build the back-end so your developers can ship code, not fix the cluster.
S-02
E-COMMERCE
Black Friday scaling, infrastructure cost optimization, CDN, cache, hot-standby database.
S-03
FINTECH & REGULATED
Operating under ISO 27001, DORA, KNF. Segmentation, logs, audit, ransomware-resistant backups.
S-04
INDUSTRY & OT
OT/IT networks, integration with PLCs and SCADA, secure bridging between the shop floor and the office.
S-05
STARTUPS
First production-grade infrastructure that won't fall over at 100× traffic.
S-06
PUBLIC SECTOR
Self-hosted within budget, data on owned hardware, compliance with public-sector policy.
06
ON-SITE ENGAGEMENT
We come to your office
for a few days — and then we automate.
Apart from our own products we also build custom solutions tailored to a specific company. The engagement starts on-site — we spend a few days at your premises, get to know your processes, watch the work as it actually happens. We draw up an action plan: what to automate, where and how, write a fixed quote — and we execute.
-
STEP 01
WE COME TO YOU
An engineer spends 2–5 days at your office, warehouse or factory. We sit with the people who do the work, observe the processes, identify where time leaks and where the duplicate work hides.
-
STEP 02
PLAN & QUOTE
We write up a document: what we saw, what to automate first, what to leave alone. Every line item with a fixed-price or hourly estimate. You decide which parts to greenlight.
-
STEP 03
WE BUILD IT
Iterative delivery with weekly demos. Integrations with your existing systems (ERP, accounting, KSeF, file shares, email). Full documentation and training for the team that takes it over.
P-01
CycKSeF
Integrated KSeF handler for Polish businesses — automated invoice processing, e-invoicing compliance and reporting built on top of the national KSeF API.
P-02
labctl
On-demand disposable Linux lab environments for trainers and instructors — spin up isolated student sandboxes in minutes, tear them down after the session.
P-03
SelfCyc
Vendor-independent self-checkout software — works with the hardware a store already has, deployed as Docker microservices.
P-04
CycSales
Loyalty cards, vouchers and promotions for store tills. Eight promotion types — from “buy 2, get the 3rd free” to card-only discounts and basket thresholds — with limits, priorities and a simulator that shows why a discount applied. Works with Netis xKasnet registers and SelfCyc self-checkouts: the till asks our server, the discount lands on the receipt.
P-05
CycIO · Cyctronics
Modular electronics from Cyctronics — load-cell (weighing), relay, I/O and audio modules that share one protocol and one SDK. Your application talks to “a scale” or “a relay”, not to a microcontroller, an ADC or a bus. In development: the first modules (LC1, LC4, Relay4, Audio Mini) are at the specification stage.
P-06
Education platforms · inf02.edu.pl
We also build e-learning platforms. Our own: inf02.edu.pl — a complete course for the Polish INF.02 vocational exam (computer systems & LAN administration): a catalogue of 52 practical tasks with step-by-step solutions and screenshots, full past exam papers, MikroTik (WinBox) and cabling (T568A/B) guides, a cheat sheet, exam strategy and a final checklist. Student accounts with progress tracking and a teacher panel for managing the class.
07
FREE TOOLS
Free tools & community.
Open infrastructure we run for the community — no signup, no SLA, just useful.
F-01
lists.cycr.xyz
Free IP blocklists — GeoBlocking for 250+ countries and 13 threat-intelligence feeds (FireHOL, Spamhaus, DShield, Tor exit nodes and more). Plain-text feeds for any firewall (iptables, nftables, pfSense, OPNsense, fail2ban…), plus ready-made MikroTik scripts. Daily updates, no signup.
F-02
BoobsOS
An atomic DevOps Linux desktop built on Fedora (bootc/OCI). The whole toolchain — containers, Kubernetes, IaC, cloud and networking — is preinstalled and configured out of the box. Shipped as a versioned image: atomic updates and one-command bootc rollback. Open-source, early access. Built and maintained by Cycrus Dynamics.
08
FREQUENTLY ASKED
Frequently asked.
Do you serve clients outside Poland?
Yes. We work remotely with clients across the European Union. Invoicing in PLN or EUR; contracts in Polish or English.
How do you bill engagements?
Three options: fixed-price project, hourly with a monthly cap, or a maintenance retainer with SLA. We pick the model that fits the work.
Do you take over existing infrastructure?
Yes — it's the most common scenario. We start with an audit, document the current state, agree on a stabilization plan, and only then make changes. Nothing gets touched "live" without a plan.
Will you sign an NDA?
Yes. We work under NDA by default. We can also operate within your access-control system and not exfiltrate any data outside the client environment.
Do you handle out-of-hours deployments and incidents?
Yes, under SLA. We run an on-call rotation and ticketing procedure. Critical deployments happen at the times of lowest production impact.
Do you only work remotely, or also on-site?
Remote by default — faster and cheaper for both sides. If the work requires presence in the data center, we travel.
I have a small budget — does it make sense to ask?
It does. We'd rather say "this isn't worth it for you, do this instead" than push an oversized solution. Even an hour of consultation can save weeks.
09
INITIATE CONTACT
Describe the problem.
We respond within
one business day.
PRIMARY CHANNEL
[email protected]
RESPONSE WINDOW
≤ 24h · BUSINESS DAYS
OPERATING REGION
PL · EUROPEAN UNION
ENGAGEMENT MODEL
PROJECT · HOURLY · RETAINER