Lead Full-Stack, Cloud & DevOps Engineer

* Conception and implementation of a Shopware 6 Web3.0 app,
which enables users to sell self-generated NFTs in their own
shop
* Development of a PHP Symfony Rest API for communication
with Shopware and Node.js and web3.js based microservices
* Containerization of the API and microservices to provide a
development environment with Docker Compose
* Setting up an AWS Relational Database Service (RDS)Aurora
database to use in the Symfony API
* Development of a frontend using VueJs and Bootstrap ,which is
displayed in the Shopware backend
* Development of the Web3 dApp using Solidity, which
automatically deploys other (NFT) smart contracts in the
Ethereum blockchain via a manufacturer pattern
* This smart contract is addressed by a dedicated Node.js
microservice, which was implemented as an AWS Lambda
Function (FaaS) using the serverless concept
* Another microservice provides the upload function of the NFT
assets, whereby the user upload is automatically stored in an
AWS Simple Storage Service (S3) bucket for backup and in a




decentralized IPFS file system for the NFT
* The entire project was deployed in the AWS Cloud (CloudWatch,
Identity and Access Management (IAM)) and the rest of the API

* In order to optimize the development process and ensure high
code quality, a CI/CD pipeline was to be developed for an
existing e-commerce full stack PHP application
* The requirements included extensive testing, which was
implemented with PHPUnit tests and Cypress.io end-to-end
testing
* The GitLab CI/CD pipeline consisted of three differentstages:
the build process, automated tests, and a final deployment
process
* In the first step, the application was containerized, building a
custom Docker image for the PHP application and the AWS
ElastiCache (Redis) cache
* These images are automatically pushed into the AWSElastic
Container Registry (ECR) .
* Setting up and provisioning of the compute resources in the
AWS Cloud by installing GitLab Runner on an ElasticCompute
Cloud (EC2) instance using Ansible
* A dedicated Amazon Virtual Private Cloud (AWS VPC) was
created for the application, creating two public AWS subnets and
an AWS Internet Gateway
* The cloud infrastructure used was provisioned using Terraform
(IaC).
* The introduction of the new process made it possible to deploy
much faster and it was ensured that bugs were detected much
earlier

* Planning and development of a hardware-based solution to
significantly optimize the effectiveness of the storage / picking
process of an e-commerce company
* A hardware put-to-light solution was developed, i.e.a trolley on
which 36 boxes (/ orders) are installed. When walking past the
storage location, LED stripes indicate the correct order / box
* A Raspberry Pi provided computing power for this project,
which was connected to a touchscreen display attached to the




car and an arcade button for operation
* To authenticate the employees using Yubikey OTP , an NFC
sensor (I2C) was also installed on the trolley.
* In order to avoid faulty shipments, a barcode scanner was also
installed, which was connected to the Raspberry via I2C as well
* The cloud-based ERP system represented the central control of
the fleet and backend, which was expanded with a REST API
using a PHP plugin.
* To keep log files in one place, all carts store them in Amazon
Simple Storage Service (S3).
* A Tkinter Python 3 GUI application was developed assoftware,
whereby the aforementioned peripherals were controlled via
libraries (nfcpy, python-barcode, GPIO) and specially developed
functions
* To streamline the manufacturing process, a pipeline was created
that automatically creates a Raspberry Pi image .
* This image consisted of a custom Linux distro, whichprovided
improved performance due to the drastic reduction in running
processes
* After switching to the new process, an increase ineffectiveness
of almost 50% was observed

* Planning and development of a customer portal / club (similar to
Amazon Prime) as an extension to the existing online shop in
order to bind customers even more closely to their brand
* Since the online store was created with PHP Symfony ,the
extension was implemented as a Symfony Bundle
* Another important goal was the possibility of blocking selected
products for all non-club customers and hiding them in the shop,
as many products were often bought up by bots due to supply
chain problems (bot detection )
* The database was addressed in the code using ORM /Doctrine
* The registration for the club took place via the registration or the
account backend, in order to store the current waiting list
centrally, an AWS Simple Queue Service (SQS) queue was used
* The cloud-based ERP system then polls the queue at regular
intervals in order to compare new registrations with the
customer data and either accepts them automatically or waits
for a review by an employee if a bot / reseller is suspected
* If the applicant is accepted, the ERP system calls up a webhook
in the Symfony app, which activates the club account
* To hide products and make others available, mainly Symfony
Event Subscribers were used

* Development of a VST software synthesizer for usewith various
DAW programs
* The synthesizer worked according to the granular synthesis
principle, which means that different sound samples were
pitched in height and lined up to produce the sound
* C++ with the JUCE Audio Library was used as the programming




language, with which the front end was also implemented with
various setting options
* A specially created Digital Signal Processing (DSP) logic was
implemented for the functionality