Cool Background Image
Profile

Hi 👋, I'm Olaru Marius-Alexandru

A passionate software developer



Education 



"Alexandru Ioan Cuza" University of Iași, România

Research Master - Computer Science

October 2025 – July 2027

Relevant Coursework

  • Advanced Chapters on Neural Networks
  • (Advanced Chapters on) Artificial Vision
  • Deep Learning with Applications in NLP
  • Programmatic Aspects for Cloud Service Engineering - AWS
  • Data Mining
  • Big Data

"Alexandru Ioan Cuza" University of Iași, România

Bachelor’s Degree in Computer Science

October 2022 – July 2025

Relevant Coursework

  • Neural Networks (DNN)
  • Machine Learning (ML)
  • Artificial Intelligence (AI)
  • PyTorch, NumPy, Python
  • Data Structures and Algorithm Design
  • Object-Oriented Programming (OOP)
  • Databases
  • Probabilities and Statistics



My personal projects 

FlappyNet

  • Time Period: dec. 2024 – ian. 2025
  • Tech Stack: Reinforcement Learning · Deep Neural Networks (DNN) · Deep Q Networks (DQN) · PyTorch · Convolutional Neural Networks (CNN) · Object-Oriented Programming (OOP) · Deep Learning
View on GitHub
Flappy Bird AI background

Description

Implemented a Deep Q-Learning algorithm with Convolutional Neural Networks to train an AI agent to play Flappy Bird. Built using PyTorch, demonstrating advanced reinforcement learning techniques.

Card background

Advanced Analytics with Apache Spark

  • Time Period: mar. 2025 - apr. 2025
  • Tech Stack: PySpark · Apache Spark · Data Science · Big Data · Problem Solving · Apache Spark ML · MLLib
View on GitHub
Advanced Analytics background

Description

Developed better the subtle concepts of Data Science using PySpark! Course supported by E.on Software Development that established the foundations of working with Big Data. This project aims to analyze the energy consumption of a group of fictitious consumers, over the course of a year, of an energy company, using the analysis techniques of the Apache Spark engine.

Card example background

Bachelor Thesis Project

  • Time Period: mar 2025 - jun 2025
  • Tech Stack: Deep Neural Networks (DNN) · PyTorch · Convolutional Neural Networks (CNN) · Encoder-Decoder · Ultra-Light-Weight · Short-Time Fourier Transform (STFT) · Object-Oriented Programming (OOP) · Deep Learning
View on GitHub
Flappy Bird AI background

Description

Deep learning (Speech Enhancement) focused on improving the clarity of human language regardless of the patient's environment! In other words, this project will adjust and adapt in real time the hearing depending on the environment with the emphasis on the quality of the conversation.

Card background

Neural Networks Laboratories

  • Time Period: sept 2024 - jan 2025
  • Tech Stack: Deep Neural Networks (DNN) · PyTorch · NumPy · Reinforcement Learning · Convolutional Neural Networks (CNN) · Deep Learning · ResNet · Object-Oriented Programming (OOP) · Git · Python (Programming Language) · Pycharm · GitHub
View on GitHub
Flappy Bird AI background

Description

I have had a deep understanding and development of neural models, from tensors and fully connected Neural Networks to the most current and used today Convolutional Neural Networks and Residual Neural Network. I am currently active in understanding the concepts of Deep Learning through the application of Transformers.

Card background



Software Skills 

PyTorch
PyTorch

Library

NumPy
NumPy

Library

Python
Python

Language

PySpark
PySpark

Library

C
C

Language

C++
C++

Language

Git
Git

Tool

MySQL
MySQL

Relational Db



Awards 

First place at the International RoboChallenge Competition

  • Category: Mini Sumo Kids
  • Time Period: 03 nov. 2019
  • Tech Stack: Embedded System · Arduino · C/C++ · Robotics
Flappy Bird AI background

Description

Robochallenge is an international robotics competition aimed especially at students who are passionate about the practice of a project, as well as those who are eager to learn more. The competition complements the theoretical part taught at the faculty, it is the way in which students can do something enjoyable and useful using the theoretical knowledge acquired. Robochallenge is a competition organized by the Faculty of Electronics, Telecommunications and Information Technology, "Politehnica" University of Bucharest, in collaboration with the Electronics Students' League (Electronics Students' League) and the ETTI Robochallenge team.


SpeedyBot 2.0 is a Mini Sumo fighting robot made of a steel chassis and front blade, five infra-red sensors (JS40F) oriented on a 180-degree front surface for detecting and attacking the opponent, two front color sensors (ML1) for timely detection of the edge of the playing surface, two 400 Rpm Micro Gearhead Dc Motors attached to one wheel each, a LiPo Battery (7.4V 2S 300Mah), all of these components being connected to the Arduino Leonardo motherboard.


FUN FACT: Immediately after winning the final, I was informed that in the last three years, 2016-2018, at the Mini-Sumo Kids competition at Robochallenge Romania, no Romanian had ranked first, second or third. I also noticed that besides me, Turkish teams had ranked second and third. It wasn’t long before I realized that most of the kits available on the market were designed by the Turks. Having their own siege of sumo robots, they had a slight advantage over everyone else.

Card background

Silver Medal at the Infomatrix International Competition

  • Category: MiniSumo
  • Time Period: 24 apr. 2019
  • Tech Stack: Embedded System · Arduino · C/C++ · Robotics
Advanced Analytics background

Description

Infomatrix is ​​an international computer science project competition, born from the desire to bring together the best IT students in the world. The competition is not only about promoting professional excellence; it also serves to promote dialogue and intercultural cooperation, by involving students and teachers from all over the world. People have a certain interest in robotics and it is growing considerably. In order to make the new generation more curious and passionate about robotics, several competitions have been organized in different countries around the world. Infomatrix invites the best minds from all over the world to compete in the Robotics category. Infomatrix has 7 subcategories for Robotics. Two of them are called Lego Mini-Sumo and Mini-Sumo.


SpeedyBot 1.0 is a Mini Sumo fighting robot made of a steel chassis and front blade, five infra-red sensors (JS40F) oriented on a 180-degree front surface for detecting and attacking the opponent, two front color sensors (ML1) for timely detection of the edge of the playing surface, two 400 Rpm Micro Gearhead Dc Motors attached to one wheel each, a LiPo Battery (7.4V 2S 300Mah), all of these components being connected to the Arduino Leonardo motherboard.


The Mini-Sumo competition consists of two autonomous robots competing and trying to push their opponent out of the field. The playing field is round, and the inner surface is painted black and is bordered by a white border. Two robots compete in each battle. A battle consists of three rounds.

Card example background

Silver Medal at the Infomatrix International Competition

  • Category: Robotics LEGO Sumo
  • Time Period: 14 may 2018
  • Tech Stack: Embedded System · Arduino · C/C++ · Robotics
Advanced Analytics background

Description

The same competition as above, but the category is different. Different strictly in terms of combat robots. These are designed this time from mechanical LEGO pieces and made up of sensors and motors that offer exactly the same functionalities as an Arduino robot! Two infrared sensors for detecting the enemy, two color sensors for detecting the combat surface, two motors attached to the wheels, as well as an attack motor, but not lastly the programming brick of the component parts.

Card example background

Bronze Medal at the Infomatrix International Competition

  • Category: Robotics LEGO Sumo
  • Time Period: 01 may 2017
  • Tech Stack: Embedded System · C/C++ · Robotics
Flappy Bird AI background

Description

InfoMatrix promotes multiculturalism and understanding between people, this is one of the main goals of this beautiful international competition. Young people from Europe, Asia, America, Africa, we all met here to compete and learn more from each other!


The Hola robot is a LEGO combat robot built and programmed in the C/C++ programming language. It is equipped with Mindstorms EV3 equipment, consisting of two infrared sensors for detecting and attacking the enemy placed in front of it, two color sensors to detect the edge of the combat surface in advance, two speed motors each equipped with a wheel, an attack motor that helps Hola push opponents more easily through a propeller and the motherboard that manages each component.


The Lego Mini-Sumo competition consists of two autonomous robots competing and trying to push their opponent out of the field. The playing field is round, and the inner surface is painted black and is bordered by a white border. Two robots compete in each battle. A battle consists of three rounds.

Card background

The finalist alliance in the FIRST Tech Challenge International Competition

  • Time Period: 20 apr. 2019
  • Tech Stack: Embedded System · Android · Robotics
Flappy Bird AI background

Description

FIRST Tech Challenge (FTC), formerly known as FIRST Vex Challenge, is a robotics competition for students in grades 7-12 to compete head-to-head by designing, building, and programming a robot to compete in an alliance format against other teams. FIRST Tech Challenge is one of four major robotics programs organized by FIRST, the other three programs including FIRST Lego League Explore, FIRST Lego League Challenge, and FIRST Robotics Competition.


The robot kit is Android-based and programmed using Java, the Blocks programming interface, or other Android programming systems. Teams, with the guidance of coaches, mentors, and volunteers, must develop a strategy and build robots based on innovative and sound engineering principles. Awards are given for robot performance, as well as community development, design, and other real-world achievements.


In the 2016-2017 school year, I was a member of the AQUILA Team (RO 040) from the “Petru Rareș” National College, which was the first team from Neamț County to participate in this prestigious competition. In the following years, several high schools in the county established teams that participated in the FTC competition. Later, after becoming a student at the National College of Informatics in Piatra Neamț, I joined the CYB3RG0DS team (RO 085) where I participated in all the activities of the FTC competition.

Card background



Volunteering

Association of Computer Science Students (ASII)IT Department Member

August 2022 – January 2025

Iași, România

  • Led the organization of the national FIICode competition, showing a spirit for coordination, rigor, and planning.
  • Expanded event to international audiences (+25%), secured BitDefender as an exclusive sponsor for the Capture The Flag area. Partnered with companies like Amazon, HEITS Digital, XWiki, Bytex, and Profidata.
  • Contributed to the FIICode website using React. Gained hands-on experience with Git and GitHub.
  • Selected among 15 students (from 200+) by GETA, a selective training division within ASII.
  • Delivered training sessions on Git, GitHub, React, and public speaking. Represented FIICode through high school presentations.



FAQ's 



Do you want to stay in touch? 

Here are my contact details:

Email

marius.alexandru.olaru@gmail.com

Phone Number

+40 737 421 575