Hello, my name is
Alexander Helsing
Software Engineer
I am a passionate software engineering student currently in my third year at the University of Gothenburg and Chalmers University of Technology. With a strong foundation in software development, I am dedicated to exploring and mastering the diverse fields within computer science. My journey has equipped me with hands-on experience in website development, embedded systems, and distributed systems. Here, you'll find insights into my projects, experiences, and skills I've gained during my years studying.
Skills & Experience
Frontend
- HTML
- CSS
- Tailwind
- Next.js
- React.js
- Bootstrap
Backend
- PostgreSQL
- MongoDB
- Redis
- ASP.NET Core
- Entity Framework Core
- Springboot
- Node.js
- Express.js
- REST API's
Programming Languages
- JavaScript
- TypeScript
- Java
- C#
- C++
- Python
- SQL
Technologies & Other
- Docker
- Git
- Postman
- OpenCV
- Torch
- Transformers
- Machine Learning
- Datascraping
- MQTT
- UML
- Trello
- Jira
- Linux
Projects
Dentist Application
TypeScript | JavaScript | Java | Express.js | Node.js | REST | Vue.js | Tailwind | MongoDB | Redis | WebSockets | MQTT
Our teeth are invaluable, yet we often overlook their care until issues arise. Finding a dentist in Sweden can be challenging due to high demand and limited availability. To simplify this process, we have developed a dentist app. Users can search, book, and cancel appointments in their city of choice, with clinics conveniently displayed on a map. Our app's microservices architecture allows for easy scalability, ensuring reliable service as demand grows.
Partiprogram.nu
Next.js | React | TypeScript | Tailwind
Partiprogram.nu är en webbplats jag skapade för att samla information om Sveriges riksdagspartier på ett lättillgängligt sätt. Projektet föddes ur min egen frustration över hur svårt och tidskrävande det kan vara att hitta information om alla partier samlat på en plats. På sidan kan besökare hitta korta beskrivningar av varje partis ideologi samt deras fullständiga partiprogram. Målet med Partiprogram.nu är att stärka medborgarnas kunskap om de politiska alternativen i Sverige och underlätta för välgrundade beslut inför val.
WorkWell AI HR Tool
NLP | Transformers | Deep Learning | Machine Learning | Next.js | FastAPI | Python | PostgresQL | SQLite
WorkWell is a web-based platform used by employees to answer daily surveys about their well-being, receiving a mental health prediction in return. Employees can also input free-text statements, and a transformer-based AI model interprets and classifies this data. A pipeline exists for the NLP model, allowing it to learn from newly submitted employee responses, and thorough testing is performed before each redeployment. There is also a rollback function enabling reversion to previously deployed models at any time. There is also a dashboard for HR where they get statistics over their department and the employees wellbeing.
Jzgolv.se
Next.js | React | TypeScript | Tailwind
I developed a user-friendly website for a local flooring company, designed to effectively showcase their services, contact information, and a selection of past projects. The site features clean navigation and high-quality visuals to engage potential customers. In addition to the design and development, I also managed the setup and hosting on a Virtual Private Server (VPS) to ensure reliable performance, security, and room for future growth.
AntiqueGuessr
C# | Asp.net Core | Entity Framework | Python | Datascraping | BeautifulSoup | Next.js | TypeScript | Tailwind
AntiqueGuessr is an interactive web game where users test their knowledge of vintage and antique items. It features two games: "Guess the Price", where players guess the value of vintage items, and "Bytt är Bytt" inspired by the Swedish game show with the same name. The backend is built with C# and ASP.NET Core, using Entity Framework and PostgreSQL for the database. The frontend is developed with Next.js and TypeScript. A Python web scraper populates the database with items. The project showcases full-stack development, integrating complex queries, and data scraping.
Work in progress: ChatBot RAG AI
NLP | Transformers | Locally Hosted AI | Springboot | Next.js | FastAPI | Java | Python
I am developing a locally hosted AI model with Retrieval-Augmented Generation (RAG) capabilities. The system allows users to ask questions about their orders, company policies, and personal profiles. By leveraging embeddings stored in a database, the AI can retrieve relevant information efficiently and provide accurate, context-aware responses. This approach ensures data security and enhances the user experience by integrating structured database queries with AI-driven insights. The AI model that the user chats with is a Llama3.2:3b model, and the embeddings are created by a embedding-generative model called "nomic-embed-text". The data is saved in a PostgreSQL database with PGVector support.
Autonomous Car
C++ | OpenCV | Machine Learning
This project showcases the development of autonomous driving software for a miniature car using computer vision and machine learning. The computer vision approach, utilizing OpenCV, tracks cones to determine steering directions, while the machine learning approach employs a Random Forest model to predict steering angles based on sensor data. Both methods were designed, implemented, and tested using data from various sensors on an RC vehicle.
LlavaStream
C++ | OpenCV | Llava | Curl
Developed a C++ application integrating the Llava AI model with OpenCV to analyze live video feeds in real-time. The system processes each frame independently, using a stateless architecture that captures frames, encodes them in Base64, and sends them via cURL to the Llava API for immediate content analysis. Displays responses as overlay text on the video, showcasing seamless AI and video processing integration with efficient API communication.
Assignment Ally
JavaScript | Node.js | Express.js | Vue.js | BootStrap
Our app streamlines assignment management by allowing you to create dedicated groups for each project. Within these groups, you can conveniently schedule meetings, either via Zoom or in campus rooms. Tasks can be assigned to group members and prioritized with labels. Each group stores all necessary resources, such as diagrams, assignment PDFs, and communication links. The user dashboard provides an overview of all tasks and meetings across all groups.
Seeed Sentinel Security System
Java | C++ | Arduino | Android Studio | XML | MQTT | IoT | MongoDB
The SeeedSentinel is a security system designed to enhance home and business protection. Utilizing sensors like infrared motion detectors and sound sensors, the system detects potential intrusions and alerts owners via an Android app. This project aims to offer peace of mind by providing a reliable security layer against burglaries and break-ins. Users can activate and manage the system through a WIO terminal, entering a password to deactivate alarms before the system switches to 'Intruder mode' and notifies them of any detected threats.
NASA Hackathon - GeoNest
Python | React | FastAPI | Machine Learning
Our project for the Space Apps Challenge 2024 in Gothenburg is a 'Community Mapping' app designed to help users find ideal living locations based on environmental data from NASA. Users set preferences and prioritize factors like air pollution, temperature, vegetation, and population density. A machine learning model processes these inputs to identify the most suitable location globally. This tool not only supports relocation decisions but can also assist in identifying optimal locations for facilities like factories. Future plans include expanding the model to consider additional data, such as GDP, soil quality, and water availability.
Contact
I'm always open to connecting and exploring new opportunities! Whether you're interested in collaborating on a project, discussing a potential partnership, or just networking, I'd love to hear from you. Feel free to reach out through the links below, and let's see how we can work together. Looking forward to connecting!