Engineering experience:
Languages:
Building back-end API for one of the world's largest VPN service - Proton VPN - which operates as the middle layer between client apps and the servers. My responsibilities span the full project lifecycle, from initial planning to production release, ensuring smooth execution across all stages. This means communication across different teams and with different stakeholders, writing design documents that outline proposed architecture together with considered alternatives, roll out/roll back plans, testing, monitoring, and the implementation itself (PHP, MySQL). Beyond the technical work, I also take an active role in the company's hiring process by conducting technical inteviews and reviewing candidates' take-home assignments as well as mentoring more junior colleagues.
Deploying and evaluating open-source groupware (i.e. email, calendar, etc.) solution for an organization with 40 000 mailboxes. The tasks carried out in this project were a mix of different parts of the DevOps world, e.g. development of migration processes (Bash, PowerShell, Python) and deployment of the infrastructure (Puppet, OpenStack), while maintaining the legacy email system.
Designing, implementing, and maintaining API for SimplySo's licensing system as well as the company's core product. The technologies used were C# and MSSQL for the application development and OAuth 2.0 framework + OpenID Connect for authorization and authentication implemented by Microsoft Azure.
Implementing front-end solution in HTML5, CSS and JavaScript (jQuery, AngularJS); developing C# Web API. The project was a web-based single page application for optimizing logistics and transportation of goods. The main task was implementing front-end features using AngularJS, however, marginal development in C# was involved as well.
I have obtained this degree with a specialization in Efficient and Intelligent Software. My thesis project focused on identifying users' location based on the language features used in their Twitter posts. Through the project, I deepened my NLP/AI knowledge and practiced researching relevant literature.
Exchange semester at McGill University, Montreal as part of my M.Sc. studies at DTU with a focus on machine learning and natural language processing which inspired me to pursue related topic in my thesis project.
The degree was practice-oriented which prepared me well for work in the industry as well as working in teams. I graduated with the highest grade for my bachelor thesis for which I also received a prize. Additionally, I was invited to give a talk on How to be a successful ICT student.
Developing chat module that can be integrated on any CERN website that uses their CMS. The chat is integrated with Lync/Skype for Business and therefore its main feature is that it allows users on the web to interact with other users directly even from environments that are not natively supported by Microsoft.
A classifier built on fastText that can identify the author of messages sent between two people. The classifier was tested on Facebook message data covering 8 years of conversations between me and my friend. After preprocessing, the dataset contained 130 000 messages and the classifier had an accuracy of 78.7%. Part of the project was to build an interactive web application where both of us could compete against the model by trying to guess who wrote which sentence. Our combined accuracy was 72% (measured on 1 000 samples) which is lower than the accuracy of the model which was quite surprising considering we are the authors of all of the text in the corpus.
... such as task management system, personal finance system, link sharing service, etc. Contact me if you'd like to see these projects, explore the live examples or if you'd simply like to ask more questions about them.