Sven "nosey" Tajz

Software engineer

LinkedIn LogoLinkedIn Logo

About

Hi, I’m Sven, a software engineer with a passion for building performant full-stack web applications. I love diving into the entire development ecosystem to create seamless user experiences.

const industryExperience = {

Software Engineering Intern - Kantun Devs (Rijeka) :
June - Present

{
  • Built a fully functional full-stack in-house EMS application. Built using NextJS, ExpressJS, Knex and Mysql.
  • Rolled custom error handling for efficient bug fixing
  • Refactored and improved code based on feedback from my mentors
NextJS
MySQL
ExpressJS
KnexJS
};

Data Engineer (Part-time) - Company name under NDA :
January - March 2024

{
  • Tasked with preparing and cleaning data from PDF files.
  • Wrote a script that scans the necessary data from a large corpus of data, processes it, cleans it and exports all data in to an Excel file.
Python
Pandas
};
};
const education = {

Masters degree, Faculty of Informatics and digital technologies, Rijeka :
October 2022 - September 2024

{
  • Created web apps in a team environment with other students utilizing AGILE methodology (Jira ticketing, organizing sprints etc.)
  • Learned a lot about machine learning and Large Language Models (LLMs)
  • Learned a lot about managing and working in a team environment
  • Learned how to implement nature inspired algorithms to solve complex problems (Ant colony algorithm, Evolution simulation algorithm etc.)
  • Published and defended my thesis Personalized Travel Assistant Based on Retrieval Augmented Generation with Large Language Models
};

Bachelors degree, Faculty of Informatics and digital technologies, Rijeka :
January - March 2024

{
  • Learned about the basics of programming, how computers work, data algorithms etc.
  • Learned about discrete math, numeric math, combinatorics and other complex math disciplines
  • Written my undergraduate thesis: ”Reading, processing and storing JSON in web applications
};
};
const projects = {
link svg icon
project image

TravelTalk

githublogoRepository link
TravelTalk is a SPA where users can create accounts and share their travels with other users. It uses Googles geocode API to convert posts to coordinates and show it on the map. Implemented a RAG-extended travel assistant for real-time user support.
link svg icon
project image

NetworkMotif

githublogoRepository link
For the assignment I used the NetworkX python library to generate random directed graphs, and developed functions that find N-sized subgraphs, which are then classified as network motifs.
link svg icon
project image

PricePrediction

githublogoRepository link
Used a open-source Kaggle to fully clean and prepare the dataset for training. I used models such as Linear Regression, Random Forrest and Extra Trees Regressor to train and test the model. Visualized all of the data and results, and saved the model so I could build the API for my TravelTalk project.
};