In data we trust!

Profile photo
Hi, I’m Arpad Dusa, a data scientist with a Master's degree in applied/nano physics, currently focused on exploring machine learning and deep learning. I am passionate about turning data into actionable insights that empower businesses to make informed, data-driven decisions. With a strong foundation in Python, I specialize in machine learning, computer vision, and natural language processing. I also have expertise in MLOps, focusing on building scalable data pipelines and deploying machine learning models into production, ensuring efficiency and reliability.

This is my personal portfolio webpage, where I showcase my projects and share insights on data science topics and books that I find interesting.

Currently Working On:

Participating in an Omdena project (weather based crop yield predictions for small scale farmers in Kitwe, Zambia)

Learning Journey:

Currently learning model deployment in AWS and Google Cloud

Tools I Use:

Python, scikit-learn, Pytorch, Docker, Weights & Biases

When I’m not wrangling data...

I love diving into sci-fi, mythology, and futurology reads, and I'm equally passionate about hitting the road on my bike or going for a run.

Our DSR Berlin final project was the highlight of the bootcamp. We developed DanceBits, an innovative machine learning application that automatically segments choreography videos into individual dance moves. Our goal was to create an interactive platform for learning and analyzing dance, making dance education more accessible and engaging.

PyTorch Weights & Biases Docker FastAPI React

This was my first project, so it is a special one for me. Not only because it was the first, but also because it's a teaching app, and I am passionate about education and teaching (I was tutoring elementary and high school pupils in math and physics, also did talent nurturing in math). It was a project in Techlabs Berlin, AI track. We developed an RAG application where you can upload a pdf/text document and ask questions about it. An another model was trained to ask questions about the uploaded text. The final feature was that the pupil could test the knowledge in a quiz.

LLM finetuning RAG Haystack Flask

Latest Blog Posts

venv, conda, pipenv or poetry?

Managing Python projects can be tricky without the right tools. In this post, I’ll break down the basics of venv, conda, pipenv, and poetry, designed to help with virtual environments and dependencies. You’ll learn how each tool works, plus get practical tips and tricks to make your workflow smoother. Whether you're new to Python virtual environments and dependencies, or just need a refresher, this post will help you choose the right tool for your needs. This is the first part of a 3-part series, so stay tuned for deeper dives into these tools!

Dependency management Virtual environments Anaconda pipenv Poetry

The Singularity is Near by Ray Kurzweil

Kurzweil in this book explores the concept of technological singularity, a future point where artificial intelligence surpasses human intelligence, leading to rapid advancements in technology and human evolution. Kurzweil argues that exponential growth in computing power, biotechnology, and nanotechnology will merge human and machine, resulting in radical changes to society, including immortality and the ability to enhance human capabilities. He predicts this will occur by 2045, and discusses both the benefits and potential risks of such transformation.

books I like Ray Kurzweil AI technological singularity

Get in Touch

Email:

LinkedIn:

GitHub: