Troels Kaldau
Software Developer with a focus on end-to-end mobile applications
Troels Kaldau
Software Developer with a focus on end-to-end mobile applications
August 2023 - September 2024
TL;DR
As CTO of MyEpi, a MedTech startup focused on epilepsy management, I led the technical transition from Firebase to a NestJS/MongoDB backend, implemented systematic feature development, and ensured smooth deployment workflows. I gained hands-on experience in backend migrations, stakeholder collaboration, and feature ideation. I also presented technical solutions to stakeholders like Pfizer and Danish hospitals and participated in investor negotiations. Despite the company's closure in 2024, this role accelerated my growth in technical leadership, strategic planning, and cross-functional collaboration.
MyEpi is a MedTech startup founded in 2021 by Jakob Junker. The idea for the company was an application to help people affected by epilepsy manage their condition.
MyEpi website (outdated)
Before I joined the company, the product consisted of a mobile application that allowed users to log their seizures, as well as an Apple Watch application that detected seizures and alerted the user's emergency contacts. The algorithm for detecting seizures was commissioned from a third party and used accelerometer data from the Apple Watch, based on a study published in 2017 (Link).
I joined the company in 2023 as a co-founder and was hired as the CTO through the company's acceptance into Innofounder, a Danish accelerator program. As the sole technical employee at MyEpi, my role was to ideate, design, develop, and deploy new features, as well as maintain the existing codebase.
I started my work by writing a "Product Opportunity Analysis" - an overview of all the ways to increase the company's value. It divided the opportunities into value categories and defined the stakeholders, short-term value, long-term value, potential, waste, and more. Despite the superficial nature of the analysis, it ended up being the guiding document for much of the company's forward development.
The system was originally based on a Flutter application with a Firebase backend, and the Apple Watch application was written in Swift. My first major task for the company was to migrate the Firebase backend to a NestJS backend with a MongoDB database. This was done using a script to extract data from Firebase Storage and restructure it to fit the MongoDB format. It was not possible to migrate users' authentication credentials, so users were asked to reset their passwords upon first login. The switch was made with ample warning to users, and no data was lost in the process.
My work mainly consisted of refactoring and improving the existing system and developing new features. When developing features, I followed a methodical approach. I would first ideate the new feature and present it to the startup's stakeholders. This included an overview of the feature, a rough cost estimate, and a value proposition. If the feature was accepted, I created a requirement document detailing the work. The feature was then implemented, tested, and deployed. I relied on endpoint testing for the backend and user testing for the frontend before release. Examples of features implemented include improved contact management and an enhanced signup flow:
The startup was funded by Innofounder from August 2023 to August 2024. During this period, I participated in numerous meetings with stakeholders, such as presenting our solution to Pfizer Denmark and various hospitals across Denmark. At the end of the funding period, we had interest from multiple hospitals in implementing our solution for a pilot project, but we had not secured revenue streams. We therefore sought investor funding, and I took part in multiple stages, including presenting our solution to potential investors, negotiating terms with new and existing investors, coordinating with lawyers, and more. This was well outside my field of expertise, but I learned a lot from the experience. Unfortunately, the funding round was unsuccessful, and negotiations broke down in September 2024. In October 2024, the company was closed down.
I entered the company knowing that a successful outcome was unlikely - that is the nature of startups. I did not join the company expecting a large payout but because I knew it would allow me to learn at a fast pace and expose me to many aspects of business that I would otherwise be shielded from. I am grateful for the experience, and I am proud of the work I did at MyEpi.
Also Read:
Project
MyEpi Full System Migration
Led the migration of all systems at MyEpi, ensuring no downtime, and took control of infrastructure and costs
Project
gRPC Performance Analysis
Optimized a real-time system using gRPC, achieving low latency and high scalability for 500+ concurrent clients for my bachelor's thesis