Błażej Dubiel

Software Engineer — Freelance

JAN 2026 – Ongoing

• Brought a prototype clarifier to life for the chemistry lab of a beet sugar factory. The scope included Python-based control software on a Raspberry Pi 5 that regulates chemical solution concentration and temperature in real time, integrating IO-Link sensors (IFM AL1340) and digital I/O (ICP-DAS ET-7245) with a Qt-based HMI. End to end and by myself.
• This website.

Python, PyQt6, Linux CLI, Claudde Code

Software Engineer — BOIKON

JAN 2023 – Ongoing · Leek, Netherlands

• One of the lead engineers behind the track-and-trace software that keeps tobacco production compliant with the traceability requirements of the EU Tobacco Products Directive.
• Developing, deploying and maintaining software across the whole chain: data applications that aggregate and process production data, control logic for industrial machines, HMI design and vision system configuration.
• Led several projects that reshaped our software architecture and brought AI into the company in a secure, safety-compliant way. Highlights include successful design and delivery of several prototype machines, propsal and adaptation of a modular approach to our software through a shared in-house library. Recently led the adaptation of Claude Code within the software department, from the initial research and infrastructure setup to defining the Markdown-based skills and routines we now use every day.
• Staying close to clients throughout a machine's life, supporting them remotely during development and service, and on the factory floor during commissioning.
• Working daily with engineers from other departments and with product managers to turn requirements into working systems.

PLC, C#, .NET, SQL, Structured Text, WPF

AI Engineer intern — TNO

JAN 2022 – NOV 2022 · Den Hague, Netherlands

• Conducted research as part of a project entitled “Autonomous Detection and Classification of Bituminous Joints in Highway Pavement Using Two-Staged and Residual-Based Convolutional Autoencoder.
• Extracted and preprocessed intensity-based data from a laser-triangulation system and developed an end-to-end unsupervised pipeline for bituminous joint detection. My work addressed the following research areas: label sparsity, anomaly detection, feature space inhomogeneity and normalization challenges.
• Thesis outcomes received high acclaim, earning a grade of 7.5 from academia and recognition from the host company, where the proposed approach established a benchmark for future projects.

python, Docker, MATLAB

Computational Fluid Dynamics Engineer & Graphic Designer — AGH Racing Team

OCT 2018 – SEPT 2020 · Cracow, Poland

• Designed and optimized aerodynamic components for a Formula Student race car using Computational Fluid Dynamics (CFD) simulations, contributing to the 1st electric Formula Student car in Poland.
• Race car sheating design and social media graphics design.

Mechatronics Engineer intern — SISTECAUT Lda.

JULY 2019 – AUG 2019 · Aveiro, Portugal

• Manipulators programming.
• PLC programming.

MSc. Artificial Intelligence — University of Groningen

SEPT 2020 – JUNE 2024

Hollistic study of AI, including subjects such as Machine Learning, Deep Learning, Computer Vision, robotics in AI, Design of Multi-Agent systems, Natural Language Processing, Data Science as well as elective in the field of science and business.

BSc. Mechatronics Engineering — AGH University of Science and Technology (Cracow)

SEPT 2016 – JUNE 2020

Engineering program, taught in English, focused on synergy between mechanical, electrical and computer engineering for mechatronics. Concluded with a thesis on face recognition for mechatronics applications.

Erasmus+ exchange program — University of Aveiro

FEB 2019 – JULY 2019

Exchange program in Portugal during which I started my journer into AI by taking Machine Learning and Computer Vision courses.

AmPhi Labs Hackathon: H/Pack for 20230

Organised by AmPhi Labs — October 2026

Together with team Giada, I took the first place at the AmPhi Labs Pack for 2030 hackathon on sustainable packaging. The jury chose our proposal as the most innovative and tangible prototype. As the team's developer, I built a Streamlit web application that analyses a non-recyclable product and proposes the alternative with the largest financial benefit and recyclability advantage, along with a written explanation of why it suits the product owner best.

AI Safety Workshop

Organised by SAIN (Safe AI Netherlands) — Fall 2026

Governance in AI

Organised by Bluedot — Fall 2025

A course on AI governance, covering regulatory frameworks and discussing possible solutions to ethical and legal challenges.

AI Safety - technical track

Organised by SAIN — Fall 2025

A course on technical aspects of AI safety, covering technical aspects of AI safety and discussion of potential risks as well as possible solutions. Concluded with a research project entitled 'Subjective overview of the mechanistic interpretability methods for LLMs'.

ML in Poland conference (2024 edition)

Conference on AI state-of-the-art in Poland.