Field Guide to Infrastructure · Vol. I

Szymon KrawczykSite Reliability Engineer

Sp. Custos fiabilitatis (Krawczyk, 2026)

✦

A hardy, nocturnally alert specimen most often observed tending large distributed systems. Distinguished by a calm temperament under incident conditions, a strong instinct for automation, and the habit of leaving every on-call rotation quieter than he found it.

Locate the Specimen ➝ View Collected Works

Based in Legnica, Poland · Native range: Linux, Kubernetes & the public cloud

Plate I.Drawn from life
Szymon Krawczyk
Szymon Krawczyk
Adult specimen, observed in natural habitat
Plate I

Description of the Species

Habitus, temperamentum et mores

Szymon is a Site Reliability Engineer who treats production like a living ecosystem: something to be observed patiently, understood deeply, and kept in balance rather than merely kept alive.

Over eight years he has migrated from Linux systems administration into platform and reliability engineering, building the observability, automation and incident practices that let product teams ship quickly without waking anyone at 3 a.m. He writes code to remove toil, sets service level objectives that engineers and business stakeholders genuinely agree on, and runs blameless post-mortems that end in fixes rather than finger-pointing.

He is most at home where Kubernetes, Terraform and a good Grafana dashboard meet, and is known for turning noisy alerting into a handful of signals that actually matter.

Classification
Reliability · Platform · Observability
Distribution
Legnica, Poland — remote across EU & US time zones
Diet
Metrics, logs, traces, and strong black coffee
Behaviour
Calm during incidents; automates anything done twice
Conservation status
Open to new engagements
Botanical illustration of the Reliability plant, with labeled parts mapping to SRE practice FIG. 1Flos SLO(the objective) FIG. 2Folium alerta FIG. 3Folium metrica FIG. 4Bacca tracea FIG. 5Folium logae FIG. 6Caulis CI/CD FIG. 7Radix infrastructurae FIG. 8Radix automatica
Fig. 1–8. Planta fiabilis: roots of infrastructure and automation feed a stem of delivery pipelines; signals bloom as leaves, the flower is the service level objective.
Plate II

Flora of the Toolchain

Species commonly found in the specimen's habitat
No. 001
Fam. Orchestraceae

Kubernetes helmsmanii

Container orchestration
  • Kubernetes
  • Helm
  • Argo CD
  • Istio
AbundanceDominant
No. 002
Fam. Declarativae

Terraformia statica

Infrastructure as code
  • Terraform
  • Terragrunt
  • Ansible
  • Packer
AbundanceDominant
No. 003
Fam. Observatoriae

Prometheus vigilans

Metrics, logs & traces
  • Prometheus
  • Grafana
  • Loki
  • OpenTelemetry
  • Tempo
AbundanceDominant
No. 004
Fam. Nubiferae

Nimbus amazonicus

Public cloud platforms
  • AWS
  • GCP
  • EKS / GKE
  • IAM
AbundanceCommon
No. 005
Fam. Scriptoriae

Golangia pythonoides

Automation & tooling
  • Go
  • Python
  • Bash
  • Kubernetes operators
AbundanceCommon
No. 006
Fam. Responsoriae

Incidens tranquillus

Incident response & SLOs
  • SLO / error budgets
  • PagerDuty
  • Post-mortems
  • Runbooks
AbundanceDominant
No. 007
Fam. Fluminaceae

Pipelinea continua

CI/CD & progressive delivery
  • GitHub Actions
  • GitLab CI
  • Argo Rollouts
AbundanceCommon
No. 008
Fam. Persistentiae

Postgresia replicata

Stateful systems
  • PostgreSQL
  • Redis
  • Kafka
AbundanceFrequent
No. 009
Fam. Custodiae

Linux securitatis

Systems & security hygiene
  • Linux internals
  • Networking
  • Vault
  • OPA
AbundanceCommon
Observed in daily use Occasional sighting All specimens identified in the field, 2018–2026
Plate III

Specimens Collected

Selected works, pressed & preserved
Fig. 9
Specimen A · Observability

Unified Telemetry Platform

Telemetria unificata

Replaced three fragmented monitoring stacks with a single OpenTelemetry pipeline feeding Prometheus, Loki and Tempo, giving 140+ services consistent metrics, logs and traces.

  • Mean time to detect cut from 18 to 4 minutes
  • Alert volume reduced by 72% through SLO-based alerting
  • Observability costs lowered ~35% via sampling & retention tiers
OpenTelemetry · Grafana · Gocoll. 2024
Fig. 10
Specimen B · Resilience

Multi-Region Failover

Regio duplex resiliens

Designed active-passive failover for a payments API across two AWS regions, with automated DNS cut-over, replicated PostgreSQL and quarterly game-day drills.

  • RTO reduced from 2 hours to under 6 minutes
  • Sustained 99.99% availability over 18 months
  • Failover rehearsed live, without customer impact
AWS · Terraform · Route 53 · PostgreSQLcoll. 2023
Fig. 11
Specimen C · Platform

Golden-Path Platform

Semita aurea

Built an internal developer platform on Kubernetes and Argo CD: one template gives any team a service with CI/CD, dashboards, alerts and canary releases already wired in.

  • New service lead time cut from 2 weeks to 1 day
  • Adopted by 30+ product teams
  • Deployment frequency up 4× with fewer rollbacks
Kubernetes · Argo CD · Helm · Backstagecoll. 2025
Fig. 12
Specimen D · Automation

Self-Healing Operator

Medicus automaticus

Wrote a Kubernetes operator in Go that detects common failure signatures (stuck jobs, exhausted connection pools, certificate expiry) and runs the corresponding runbook automatically.

  • Eliminated ~40% of after-hours pages
  • Every action audited & posted to the incident channel
  • Released as an internal open-source project
Go · controller-runtime · Slack APIcoll. 2022
Plate IV

Field Journal & Migrations

Observations recorded over successive seasons
2023 — Present

Senior Site Reliability Engineer

Fintech scale-up · Remote

Lead SRE for the payments and platform domains, serving roughly 4 million monthly active users.

  • Introduced SLOs and error budgets across 60+ services, now used in quarterly planning
  • Designed the multi-region failover strategy and run regular game days
  • Built the golden-path developer platform; mentor four engineers on the team
2021 — 2023

Site Reliability Engineer

E-commerce marketplace · Wrocław

Joined during the move from VMs to Kubernetes and helped carry it through Black Friday peaks.

  • Migrated 90+ services to EKS with zero-downtime cut-overs
  • Wrote the self-healing operator, cutting after-hours pages by about 40%
  • Coordinated the incident-commander rotation and blameless post-mortem process
2019 — 2021

DevOps Engineer

SaaS analytics company · Hybrid
  • Codified all infrastructure in Terraform and Ansible; retired hand-built servers
  • Built CI/CD pipelines that took releases from weekly to multiple times a day
  • Rolled out the company's first Prometheus & Grafana monitoring
2017 — 2019

Linux Systems Administrator

Hosting provider · Legnica
  • Ran a fleet of 400+ Linux servers, along with networking and backups
  • Automated provisioning with Bash and Python, saving roughly a day per week
Credentials

Certifications & Education

  • Certified Kubernetes Administrator (CKA)
  • AWS Certified Solutions Architect — Professional
  • HashiCorp Certified: Terraform Associate
  • B.Sc. Computer Science
Plate V

Habitat & Where to Observe

Locality data for correspondence
Herbarium LabelCat. No. SK-2026
First name
Szymon
Surname
Krawczyk
Locality
Spacerowa 4
59871 Legnica
Poland
Telephone
+1 402 400 4716
Post
dhanpatidubianskyqatelu63@gmail.com
Det. & leg.Szymon Krawczyk
“Hope is not a strategy; but a well-tuned alert, a clear runbook and a rested on-call engineer come remarkably close.”From the specimen's field notes

Szymon is open to Site Reliability, Platform and Infrastructure roles, as well as consulting on observability, SLO adoption and incident-management programmes. He is most easily found by email; replies usually arrive within one business day.

Send a Letter ✉ Telephone ☏