Circle Image

Simon Pirschel

DevOps Specialist
Cologne/Bonn Area, Germany
Docker Kubernetes SaltStack Automation CI/CD Containers Cloud

I build next generation IT infrastructure based on container technology, cloud services and automation.

With 10+ years experience in development and operations, I combine a deep understanding of Linux systems with a broad knowledge in software engineering.


DevOps is an emerging model of product delivery that facilitates higher and faster rates of change through the optimisation of development and delivery processes. [...] As the Fourth Industrial Revolution presses upon us all, we and many in our industry believe successful Digital Transformation initiatives will need DevOps to be the driving force.

I'm here to assist you to tackle the challenges of the Digital Transformation. No matter if it's migrating an existing Software stack to a new System Architecture or platform, or building a new service from the ground up. I offer a decade of experience in Dev and Ops combined with a battle proven stack of services to build next generation infrastructure.



  • DevOps Engineer
  • Software Engineer
  • Architect



  • Continuous Integration
  • Continuous Delivery
  • Configuration Management
  • Orchestration



  • Public and Private Cloud
  • Containers
  • Virtualization



  • Load Balancing
  • High Availability
  • Web Services
  • Databases
  • Key-Value Storage


Founder & DevOps Specialist 2016 - present

Pirschel DevOps IT-Services

Chief Architect 2015 - 2016

Vice President Engineering 2013 - 2014

DevOps Engineer 2012

Co-Founder 2012 - present

Abusix, Inc

DevOps Engineer 2009 - 2011

Linux System Administrator 2008

Gameforge AG

Linux System Administrator 2004 - 2008



Computer Science Expert 2004 - 2007
Fachinformatiker Systemintegration, HHEK Bonn

Comprehensive School 1995 - 2003
Gesamtschule, Europaschule Bornheim

Elementary School 1990 - 1995
Grundschule, Wendelinus-Grundschule Sechtem


Simon Pirschel
 Bonn, Germany
 Cologne/Bonn Area, Germany

Dev Stack (TOP 5)

Python Perl JavaScript C Java

Ops Stack (TOP 5)

Debian/Ubuntu Linux SaltStack Docker Jenkins ELK Stack

DB Stack (TOP 5)

ElasticSearch Redis SophiaDB/LevelDB MySQL/MariaDB OpenLDAP

Tech Stack (TOP 5)

Stream Processing RESTful APIs Web scraping Data mining Distributed systems


Kubernetes, Docker, GitLab, Prometheus based infrastructure
Year: 2017, Duration: 5 month, Gameforge Productions GmbH, Role: Trainer, Architect & DevOps Specialist

Design and implementation of container infrastructure based on Docker and Kubernetes. Design and implementation of CI/CD processes based on GitLab. Design and implementation of a metrics monitoring system based on Prometheus. Training and support of ~60 developers and ~15 ops employees.

Docker Kubernetes GitLab Prometheus Ansible

REST API for ETL jobs and front-ends
Year: 2017, Duration: 3 month, Big TV media company Munich/Unterföhring, Role: Python Developer & Software Architect, 60% Remote

Design and implementation of a RESTful web service as API for an ETL pipeline and several web front-ends. The web service was implemented in Python with the connexion library and Flask. The API reflected a rather complex domain model with tight security requirements.

Python Connexion Flask PostgreSQL Swagger Docker Docker Swarm

ETL pipeline in Python with Luigi
Year: 2017, Duration: 2 month, Big TV media company Munich/Unterföhring, Role: Python Developer & Software Architect, 60% Remote

Design and implementation of a ETL pipeline in Python Luigi to clean and transform several data sources for big-data analysis. ETL jobs were executed in Docker containers.

Python Luigi PostgreSQL Docker

Docker Training
Year: 2017, Several customers, Role: Trainer

Two day training program for Docker and several Docker tools. Goal of the training was to bring people with Dev or Ops background and little Docker knowledge up to speed. Training included a deep dive into the Docker technology stack and hands on sessions with real projects of the teams.


Infrastructure automation and CI/CD system with SaltStack, Jenkins and Docker on AWS
Year: 2017, Duration: 2 month, Customer: Startup from Karlsruhe, Role: Architect & DevOps Specialist, 100% Remote

Concept and implementation of a fully automated container based CI/CD cloud infrastructure. Goal of the project was to implement processes and tools to enable the existing team to deploy and operate their SaaS product fully automated and tightly integrated with AWS.

SaltStack Docker Jenkins AWS CI/CD Python

Migration of an existing infrastructure to container based CI/CD public/private cloud
Year: 2016-2017, Duration: 6 month, Customer: Startup from Vienna, Role: Architect & DevOps Specialist, 100% Remote

Concept and setup of a container based CI/CD public/private cloud infrastructure. Dockerization of all existing Services (Python, Ruby, PHP, Java, Node apps). Migration of all services to containers. Host configuration, service and container orchestration with SaltStack. Continuous Integration with Jenkins. Continuous Delivery with SaltStack. API and web app loadbalancer with NGINX and AWS ELB. Docker cluster overlay network with VXLAN and etcd.

SaltStack Docker Jenkins NGINX MongoDB Cassandra RabbitMQ OpenLDAP etcd CI/CD AWS EC2 Private Cloud

System Architecture and Setup of a Slack like Chat Service
Year: 2016, Duration: 3 month, Customer: Startup from Berlin, Role: Architect & DevOps Specialist, 100% Remote

Concept and setup of a Slack like Chat Service offered as SaaS. The project had reliability as well as performance goals. AWS EC2 and Private Cloud infrastructure mix. Service and container orchestration was done with SaltStack. Several loadbalanced web applications and a MongoDB cluster in a multi datacenter setup. Services operated in a Docker overlay network provided by flannel clustered through etcd.

SaltStack Docker etcd flannel Rocket.Chat NGINX MongoDB AWS EC2 Private Cloud

Proove of concept for a container based distributed Task Queue
Year: 2016, Duration: 4 month, Customer: Big german Telco, Role: Architect & DevOps Specialist, 100% Remote

Concept and POC for a container based Distributed Task Queue to manage workflows. The Distributed Task Queue was implemented in Python using Celery. Services provided via Docker containers. RabbitMQ used as AMQP provider for Celery and ElasticSearch for event logging.

Docker Python Celery RabbitMQ ElasticSearch Private Cloud

High Performance Time Series Database for IPv4/v6 Blacklistings
Year: 2015, Duration: 3 month, Customer: Startup from Karlsruhe, Role: Architect & Software Engineer

API driven database to historical store billions of IP blacklistings across several blacklist providers through a multi dimensional key design for highly effective range queries to provide an aggregated blacklist service for network operators.

  • Store billions of entries
  • Filterable by IP, CIDR, time, time range and blacklist provider
  • Result response in under 50 msec.
PyPy SophiaDB

Network Attack Analysis & Reporting Tool
Year: 2015, Duration: 4 month, Customer: Startup from Karlsruhe, Role: Architect & DevOps Specialist

Tool for statistical analysis based on network related attacks with data coming from Spam trap and Honeypot networks.

  • Store at least 1 year of data with a daily amount of 400 million new entries
  • Near real-time data intake
  • Filterable by Date, Src. IP, Dst. IP, Src. Port, Dst. Port, Service, Protocol, Country Code, Payload, Attack Type
  • Visualize in a customizable dashboard
Python Redis JavaScript ElasticSearch Kibana

High Performance & Self-contained Network Service Honyepot
Year: 2015

Honeypot to record network based attacks on well known services like databases, remote shells, HTTP servers etc. reported near real-time to a centralized data sink.

  • Low memory footprint
  • Easy deployment
  • Security measure to avoid leak of honeypot node
  • Emulate TCP/UDP services
  • Automatic updates and control features
  • Secure and near real-time event data transport
Python gevent

Network Abuse Handling Tool
Year: 2012-2015

See: SaaS product abusehq.

Python Redis ElasticSearch OpenLDAP JavaScript Backbone.js Marionette.js Bootstrap

Pipeline oriented event processing framework
Year: 2012

Framework to build pipeline oriented work flows for data event processing through isolated pluggable components.

Python Redis

Search Engine Monitoring System
Year: 2011

Position monitoring of URLs in search engine result pages for given keywords acquired by keyword analysis of competitors and search engine suggestions. Tool was used to measure SEO activities.

Python Pyramid MySQL

Online Game Transaction Tracking System
Year: 2011

System to track and identify bot activities by gold farmers by tracking ingame transaction events.

Python CEP

Server Infrastructure & Role Management Tool
Year: 2009-2010

System to manage thousands of server instances and the application level role of the server to provision and deploy services.

Perl MySQL

User Account Load Balancing System
Year: 2009

System for web games to evenly distribute new users across instances.


SSH Key Deployment & Privilege Management Tool
Year: 2008

System to manage SSH keys, access to servers and SSH key deployment.

MySQL Perl



Kubernetes: Using containerd 1.1 without Docker

This article is about using containerd 1.1 directly with Kubernetes instead of Docker. Since 2008 with the first release of LXC and the release of Docker in 2013 a lot has happened in Linux container...

format_align_left Read Article

Docker version pinning for Kubernetes on Ubuntu

Kubernetes supports the container engine Docker only up to a specific version (currently 17.03). For that it needs to be ensured that apt wouldn’t update Docker to any unsupported version. The base installation of Docker...

format_align_left Read Article

Python 3 type checking and data validation with type hints

Type hints are one of my favorite features in Python 3. Starting with Python 3.0 PEP 3107 – Function Annotations introduced function annotations and was later enhanced by PEP 484 – Type Hints in Python...

format_align_left Read Article

Azure ActiveDirectory OAuth2 JWT Token Validation with Python

Just finished integrating Azure ActiveDirectory OAuth2 with a Python Web API using the following authentication scenario. The JWT token is requested through a web application and passed to the Web API for resource access. The...

format_align_left Read Article

How to create an Ubuntu 16.04 AWS EC2 AMI with Salt Minion pre-installed

Creating your own AMIs with AWS EC2 is pretty simple. You basically create a new EC2 instance, based on some image, modify the instance to your liking, shutdown the instance and create an AMI based...

format_align_left Read Article

Save credentials for Git HTTP/HTTPS auth with credential.helper

Recently I was annoyed by typing in my credentials for Git HTTP/HTTPS auth all the time. On my search for a solution I found the credential.helper. With the git-credential-cache Git will cache your credentials for...

format_align_left Read Article

ciso8601 vs mxDateTime vs udatetime

In my last performance comparison I compared libraries with non performance goals with datetime and udatetime, to illustrate the impact of your library choice. Arrow, Pendulum and Delorean are awesome choices when it comes to...

format_align_left Read Article

datetime vs Arrow vs Pendulum vs Delorean vs udatetime

The big alternatives to Python datetime all share similar goals. These goals are ease of use, simplicity and intelligent/user friendly API design. Awesome goals and I love those libraries for investing a lot of effort...

format_align_left Read Article

udatetime now with ultra fast PyPy support

I just finished the performance optimized pure Python implementation of my RFC3339 date-time library udatetime for PyPy and Python 3.5. The benchmark say PyPy is now officially the fastest with udatetime. Again it’s astonishing how...

format_align_left Read Article

udatetime a fast RFC3339 compliant date-time Python library

Working with date-time formats can be pretty upsetting because of the variate of different formats people can come up with. date-times are used everywhere not just only logging or meta data in database entries and...

format_align_left Read Article