At Allianz I built the base frameworks for the customer self-service, claims, broker and agent portals, which millions of customers used. From 2018 to 2024 I was VP of Engineering at Avenue I Media, a US adtech startup. While it grew from zero to $78M in annual revenue, I designed and built almost all of its platform, including the customer portal and reporting. I work on the backend and the frontend, and I tend to stay with a system for years.
I build customer, partner and internal portals, including the backend, the database and the frontend. Most portals also need reporting, which I’ve built both inside a portal and as a separate reporting platform.
If nobody looks after your portal anymore because the agency or developer who built it has moved on, I can take it over and keep working on it, from bug fixes and updates to new features. I start by reading through the code and the setup and writing down what I find, so you know where the portal stands before any larger change.
I’ve worked on a lot of old code, for example at Allianz on an integration layer that hundreds of developers had maintained over decades.
When a portal or internal app has to be replaced, I do it step by step, so the old system keeps working until the new one has taken over. At Allianz the new app used the same database as the old one, and both ran side by side until the old one was retired. At my current client I moved the company’s sites to a new platform in phases.
Portals usually have to exchange data with other systems. I’ve built APIs for partners, uploads to the APIs of six ad networks, data pipelines into Snowflake, and search on Solr and OpenSearch. That includes the less visible parts, like data models, migrations and tuning queries and indexes.
I add LLM steps to existing systems, for example bulk content generation, translation or compliance review for ads and articles. I build them like the rest of the backend, on queues and with alerting, and for compliance checks I pair the LLM review with deterministic rules.
Some companies need one person to own the technical side. I can take that on, including architecture, hiring, engineering processes and a good share of the hands-on development, plus the technical due diligence if the company is sold. I did this at Avenue I Media for six years.
When I joined, the company had no revenue and ran its operations on spreadsheets. I designed and built almost all of the platform that replaced them, including a customer portal, a reporting platform, traffic routing, a bidding engine, an ad server, two rule engines (one of them for compliance and blocking) and the revenue attribution pipeline. It scaled to 100M clicks a day.
I also hired and led a team of up to 20 engineers and set up the engineering processes from scratch. For both sales of the company I prepared the technical due diligence, which meant writing a lot of documentation, and moved systems and processes over to the new owners, down to switching the ticketing from Monday to Jira.
Python · Django · PostgreSQL · MySQL · Redshift · Redis · SQS · AWS
At Allianz I worked on the customer self-service, claims, broker and agent portals, which millions of customers used. They ran on Java/Spring Boot and Angular. I built the base frameworks for the portals, which other countries then adapted for their own markets, and a lot of the Angular components. Day to day it was new features, bug fixes and maintenance. I also built an integration that showed BIRT reports directly in one of the portals.
The portals reached the company’s business logic through an integration layer, a large Java application on Tomcat with sticky sessions. In a team of four I was responsible for the web layer of a session migration that moved live sessions off underused servers so they could be shut down. It cut infrastructure costs significantly.
Java · Spring Boot · Angular · Tomcat · BIRT
I work on the CMS behind the company’s article sites, with scheduled publishing, versioning, tags, bulk operations and article search and filtering, and on a content API that partners use to syndicate and create articles. I moved the remaining sites to a modular, plugin-based platform in a phased rollout.
I also built language-aware related-article search on OpenSearch and tuned MongoDB queries and indexes for article listing and search. Part of my frontend work there is in Nuxt.
.NET · ASP.NET MVC · Nuxt · MongoDB / DocumentDB · OpenSearch
I built the pipeline that generates articles in bulk with OpenAI through Semantic Kernel, running on asynchronous RabbitMQ workers. It also translates articles and titles and adds stock or AI-generated images. Articles that perform best get rewritten automatically.
The compliance checks I built combine deterministic rules with an LLM review of ad titles, content titles and articles. Scores are published to Kafka and alerts go to Slack. Non-compliant articles submitted by partners are blocked.
With one other engineer, I replaced a PowerBuilder project management and time-logging app with a Java/Spring Boot and Angular web app in about a year. The new app used the same DB2 instance and schemas, so old and new ran side by side until the old one was retired.
As the only engineer on it, I also built a company-wide search engine on Apache Solr that indexed several million documents from tickets, documentation, source code and help desk systems. It caught on across the company soon after launch.
2024 – today
Adtech company, current client · remote
I work on a .NET platform of around 23 microservices and 5 site applications. Besides the projects above, I rolled out a new architecture for conversion uploads to Facebook, Google, TikTok, Kwai, Bigo and Zemanta, built a Kafka-to-Snowflake ingestion pipeline in Java, and built crawlers for Google and Facebook ads and campaign automation for TikTok and Taboola.
2018 – 2024
Avenue I Media, US adtech · remote
I led engineering there from the start.
2015 – 2018
Allianz
Full-stack work on the company’s web portals and internal applications, on Java/Spring Boot and Angular. The projects are described above.
2012 – 2015
TU Wien
I organized and supported student groups in workflow and process management courses, which used Apache Camel and Activiti.
2006 – 2012
Hydrogen Research AG
I built web interfaces for embedded devices, from C on embedded Linux and microcontrollers to HTML and JavaScript in the browser, part-time alongside school and university.
Education
BSc Business Informatics, TU Wien (2013) · HTL Braunau, Electronics (2007)
Authmon monitors SSH logins. Servers send it their logs over remote syslog, and it keeps a history of logins and sessions with their geo-location. When one of your rules matches or it detects a suspicious pattern, it sends an alert.
I’ve built it alone since 2019, in Elixir/OTP with RabbitMQ, TimescaleDB and a Vue.js frontend. It’s publicly available but not officially launched. The architecture is designed for large scale, though for now I use it for my own servers and my clients’ servers.
Alerts go out by email or webhook. The web UI has login maps and session tables.
If you already have a portal, send me a link or a few lines on what it’s built with and what should change. For a new one, a short description of who will use it and what it should do is enough to start. I work remotely from Austria, in German or English.
I work through my company, MIB Digital Solutions GmbH.
Informationen gemäß § 5 ECG und § 14 UGB, Offenlegung gemäß § 25 MedienG
MIB Digital Solutions GmbHThe controller is MIB Digital Solutions GmbH (address in the Impressum). This site sets no cookies, uses no analytics or tracking, and loads no fonts, scripts or other content from third parties.
When you open the site, the hosting provider’s server automatically processes technical data such as your IP address, the date and time, the page requested and your browser’s user agent. This is necessary to deliver the site and keep it secure (Art. 6(1)(f) GDPR).
If you email me, I use your address and message only to handle your inquiry and, if we work together, to perform the contract (Art. 6(1)(b) and (f) GDPR).
The links to LinkedIn and authmon.io are plain links. Those sites receive no data from here unless you click one.
You have the right to access, rectification, erasure, restriction of processing, data portability and objection. To exercise them, email the address in the Impressum. You can also lodge a complaint with the Austrian Data Protection Authority (dsb.gv.at).