🌩️ Markus Tacker

Trash reports are created by volunteers who care about the environment.

Go to trash.fjordcleanup.org to discover more and report trash yourself.

Together, we can make the Oslo fjord and Akerselva cleaner.

I was transferred to the nRF Cloud team, and I am leading a team that is responsible for multiple services that span the entire project: accounts, billing, metrics, and frontend enablement. My work focuses heavily on migrating from a monolithic legacy code base written in TypeScript to a modern microservice architecture that enables continuous deployment and trunk based development without breaking the existing user's experience with the platform. I am primarily working with AWS serverless… Vis mer

I was transferred to the nRF Cloud team, and I am leading a team that is responsible for multiple services that span the entire project: accounts, billing, metrics, and frontend enablement. My work focuses heavily on migrating from a monolithic legacy code base written in TypeScript to a modern microservice architecture that enables continuous deployment and trunk based development without breaking the existing user's experience with the platform. I am primarily working with AWS serverless services and delivery automation through GitHub actions. Vis mindre

teamstatus.space allows distributed teams collect project status updates asynchronously and distributed to speed up their sync meetings.

I've led the effort to build a new, radically simplified out-of-box experience for Nordic's flagship cellular IoT prototyping platform, the Thingy:91 X. For the first time, all devices ship with a QR code that allows users to see real-time data published by their devices within seconds and without manually onboarding the device to nRF Cloud.

In addition I pioneered the introduction of a new sensor data protocol using LwM2M, CBOR and CoAP that significantly decreases power consumption and… Vis mer

I've led the effort to build a new, radically simplified out-of-box experience for Nordic's flagship cellular IoT prototyping platform, the Thingy:91 X. For the first time, all devices ship with a QR code that allows users to see real-time data published by their devices within seconds and without manually onboarding the device to nRF Cloud.

In addition I pioneered the introduction of a new sensor data protocol using LwM2M, CBOR and CoAP that significantly decreases power consumption and enables the support of NB-IoT for an even better compatibility with cellular IoT networks around the globe. Vis mindre

Developed a trade show demo from scratch that highlights Nordic's cellular IoT features: near real-time message transmission, ultra-low power capabilities that allows an asset tracker to be powered using a small solar cell, and alternative location services (based on cellular and WiFi cell information).

The project is continually updated and extended to support other tradeshows and our tech tours.

I was part of the team that was involved in the successful acquisition of Memfault for $120M. This involved reviewing their codebase, development principles, and interviewing key personnel.

Implemented the CoAP test server for the Developer Academy courses (https://academy.nordicsemi.com/) in Go.

Implemented the MQTT test server for the Developer Academy courses (https://academy.nordicsemi.com/).

Implemented the UDP echo test server for the Developer Academy courses (https://academy.nordicsemi.com/) in Python.

Developed and provided highly salable REST echo server used by the course participants to explore the capabilities of Nordic's nRF7 Wi-Fi chip.

For this project I am the lead engineer and architect responsible for the full-stack, end-to-end, multi-cloud implementation of a reference IoT project for an asset tracking application. The project aims to provide a concrete end-to-end sample for an IoT product so that our customers can focus on designing the best hardware for their product while providing a quick-start for them to develop a custom cloud-based solution for their product. The project currently supports Amazon Web Services, and… Vis mer

For this project I am the lead engineer and architect responsible for the full-stack, end-to-end, multi-cloud implementation of a reference IoT project for an asset tracking application. The project aims to provide a concrete end-to-end sample for an IoT product so that our customers can focus on designing the best hardware for their product while providing a quick-start for them to develop a custom cloud-based solution for their product. The project currently supports Amazon Web Services, and Microsoft Azure. It also shows third-party cloud-to-cloud integration for nRF Cloud Location Services, Memfault, and AVSystem's LwM2M device management platform. Vis mindre

Developed a completely new out-of-box experience for this collaboration between Avnet, Exeger and Nordic Semiconductor that allows users to retrieve real-time data from their devices within seconds after scanning a QR code.

I managed and mentored 6 students during a full-time summer project to build an interactive, real-time strategy robot fighting game built using Nordic wireless solutions (Bluetooth Mesh for the robots, and LTE-m for the cloud connectivity). Besides providing the project idea and scope I also was involved in the training the students in various cloud-related technologies to build the web application that controls the game.

Proof of concept of an LWM2M Gateway in front of AWS IoT using Eclipse Leshan.

This project was developed to determine NAT timeouts in different combinations of SIM card provider and network operator, testing UDP and TCP NAT timeouts.

I lead the effort to certify the nRF9160 for the Azure Certified for IoT device catalog
in the Azure Marketplace.

I've built a demo application for embedded world 2020 that highlights the UDP and NB-IoT capabilities of the nRF9160.

For this project I was the lead backend engineer and architect as part of the IoT Lab which is a distributed team based in Portland, US and Trondheim, Norway. I was responsible for the concept and implementation of major feature for nrfcloud.com, Nordics cloud-based IoT prototyping platform, as well as for the operation of the platform which is built using AWS in a serverless fashion.

My main tasks besides architecture design and implementation were improving the developer experience… Vis mer

For this project I was the lead backend engineer and architect as part of the IoT Lab which is a distributed team based in Portland, US and Trondheim, Norway. I was responsible for the concept and implementation of major feature for nrfcloud.com, Nordics cloud-based IoT prototyping platform, as well as for the operation of the platform which is built using AWS in a serverless fashion.

My main tasks besides architecture design and implementation were improving the developer experience and reliability of deployments by introducing and implementing CI/CD principles and streamlining the release process.

In 2018 I lead the development of the first version of the nRF Connect for Cloud Device API which enables our customers to interact with their devices and easily react on data sent by devices using their own business platforms without needing to know too much about the intricacies of IoT hosting by using a simple RESTful API. This involved spending a substantial amount of time doing market research in order to understand what capabilities our customers most likely would need from an IoT cloud service when they start to iterate on developing their IoT product with Nordic hardware. I always strive to understand the entire value chain, and really digging deep into a topic is importing to build a solid foundation for entirely new service offerings.

Read an introduction about the project here:
https://www.nordicsemi.com/eng/News/ULP-Wireless-Update/Dedicated-IoT-Lab-prepares-for-consumer-boom Vis mindre

Review of technical implementation of a SaaS platform, its overall architecture as well as business capabilities. Delivered on-site evaluation and written report.

I’ve built the backend for the solutions using event sourcing and the shared core components have been open-sourced as RHeactor on GitHub. Those components are continuously tested via Travis and follow widely-accepted rules for formatting source code and releasing packages which enabled us to effortless share code within our projects.
The backends are running on AWS and leveraging their IaaS products ElasticBeanstalk and Lambda which provide virtual machines that are managed and maintained… Vis mer

I’ve built the backend for the solutions using event sourcing and the shared core components have been open-sourced as RHeactor on GitHub. Those components are continuously tested via Travis and follow widely-accepted rules for formatting source code and releasing packages which enabled us to effortless share code within our projects.
The backends are running on AWS and leveraging their IaaS products ElasticBeanstalk and Lambda which provide virtual machines that are managed and maintained by Amazon. Choosing this approach enabled us to focus on building the products and not on buying, provisioning, updating and monitoring servers. It also provides the foundation to scale quickly for more users. RHeactor can be launched stateless and all components communicate via an event bus. The event-based data model is also ideally suited for distributed architectures. The websites are deployed to static hosting platforms, either on AWS S3 or on GitHub pages.
Updates to our products can be deployed automatically. Push to deploy is realized using Travis CI and make. For quickly iteration on new features I’ve implemented the ability to test features for certain users by using feature switches.
Code is written in a TDD manner, using BDD tests against the public API, and unit tests for libraries.
I implemented an agile process for product development using a kanban based approach in Trello which provides a clear tactical overview of what is currently happening in the IT department and on what tasks our contractors are working on at the moment. Vis mindre

Fintura had launched a closed-beta in one of our states and my task was ensure a smooth German-wide launch in July. I defined certain goals that were critical for me like the ability to push-to-deploy, high availability, service-level monitoring, and disaster recovery plans. We implemented them for the launch. Besides these organisational tasks I worked on the website and reworked it completely, moving it from a collection of .php files to Jekyll. By launch time I also worked on the core… Vis mer

Fintura had launched a closed-beta in one of our states and my task was ensure a smooth German-wide launch in July. I defined certain goals that were critical for me like the ability to push-to-deploy, high availability, service-level monitoring, and disaster recovery plans. We implemented them for the launch. Besides these organisational tasks I worked on the website and reworked it completely, moving it from a collection of .php files to Jekyll. By launch time I also worked on the core Node.js application which powers the loan comparison process.

When I joined Fintura, the management’s way of managing the tech team was to walk in the room were the developers are sitting and assigning tasks directly. Being a Kanban fan I’ve established a process that integrates strategic goals, features and bug fixes. This was quickly picked up and proved to be have a great positive impact on the stress level for the developers.

A second major project was the shift from cloud hosted services (Google Apps, Dropbox) for our office tools to self-hosted systems. For us it was important to have our customer data completely under our control. This was a major effort and required a lot energy from the tech team but also from the rest of our staff because the self-hosted tools are not as convenient as the Google products. But recent developments (the EU Court of Justice declaring Safe Harbour to be invalid) showed that we made a good decision.

Apart from that I defined the future team structure (a 10 person tech team) and started to hire for that. Fintura’s goal by that time was to raise a $5M series A with a $25M valuation in August with an IPO for a $500M valuation planned for 2020. Vis mindre

When I joined dotHIV they had just signed the TLD contract with ICANN and where due to launch into general availability by August 2014. Back then the team consisted of a handful contributors which since then grew to a dozen people. During my time at dotHIV I was mostly the only developer on the dotHIV staff but I had to constantly manage contributions from supporting individuals and also the marketing agencies thjnk and SinnerSchrader.

In the month up to the launch I’ve prepared our… Vis mer

When I joined dotHIV they had just signed the TLD contract with ICANN and where due to launch into general availability by August 2014. Back then the team consisted of a handful contributors which since then grew to a dozen people. During my time at dotHIV I was mostly the only developer on the dotHIV staff but I had to constantly manage contributions from supporting individuals and also the marketing agencies thjnk and SinnerSchrader.

In the month up to the launch I’ve prepared our infrastructure for the general availability of the TLD. The big launch was in August 2014. We not only had to prepare for a huge increase in traffic which I did by employing strong HTTP caching capabilities on our website but also integrate our technical service providers data warehouse into our customer journey and our own management tools. I also had to rewrite our most critical component, the click-counter which displays a thank-you-note when a user visits a dotHIV domain, to ensure its reliability. I chose a mixture of technologies to ensure quick development and maintenance time. The website is powered by a Symfony2 app with Contentful as the CMS. The business components related to our domains are written as micro-services in Go. Both are self-hosted. The click-counter is written in Python and runs on the Google AppEngine. All components communicate via a RESTful pull API.

During that time I also made it my job to improve the collaboration in our team and introduced Trello, Dailies and Weeklies which enabled us as a remote team to bond and deliver results quickly with the least amount of communication problems. Vis mindre