AI ENGINEER / SOFTWARE ENGINEER / PROFESSIONAL MAKER
I build software
with AI.
I have more than ten years of experience in software and engineering, including more than six in industrial computer vision.
More about my work
I'm a professional maker. I use AI across research, design and development to turn ideas into working software. I set the direction and take responsibility for the result.
Two aluminium sections and two plastic strips. Compare their shapes with CAD and check the four joining locations. Generic illustration; no measured results or joint-strength test.
THE PROBLEM
The existing vision system needed more reliable checks of assembled aluminium profiles and clearer explanations when an inspection failed.
MY CONTRIBUTION
I corrected processing errors, tuned settings for individual profiles and built web views for operators, managers and engineering analysis.
EVALUATION
I evaluated the changes on archived images. A pilot on new production images and final installation checks are still needed.
Two cameras inspect the top and bottom surfaces separately. For example, the top camera may see a scratch while the bottom camera sees a dent. Each defect is classified by type and severity from 1 to 3. Severity 3 on either surface means scrap the strip. Generic illustration: the equipment and defects are fictional, not an Aperam installation or an inspection result.
THE PROBLEM
Find millimetric surface defects in high-resolution images of coils.
THE APPROACH
I built an image-processing algorithm in Python using OpenCV and PIL.
OUTCOME
I helped roll out the inspection solution across multiple plants, improving inspection reliability.
03Professional • Internal project • Jan 2026-present • Active development
An engineer compares the proposed tags with the drawing, corrects them and exports reviewed data. Fictional drawing and sample inventory; no client documents or application screenshots are shown. Extraction still requires engineer review. Time savings and error reduction remain unverified.
THE PROBLEM
Engineers spend hours reading complex P&IDs and checking tag inventories. Changes to the drawings add more work and opportunities for mistakes.
THE APPROACH
I use AI to prepare the first inventory. Engineers check it against the drawing, correct missing or incorrect information, and export their reviewed data.
OUTCOME
I am developing the extraction, correction and export workflow. Time savings and error reduction remain unverified, and validation across drawing vendors is limited.
Illustrative workflow, not a product screenshot. One Project links a human-accepted Specification to planning, coding-agent implementation, verification and recorded evidence. Story closure, Sprint review, closure and learning are separate steps. AgileForge rules choose eligible actions; ADK generates candidates. AI responses still vary and need review. Independent open-source project.
THE PROBLEM
Keep AI agents working from requirements that a person has approved.
THE APPROACH
I built a workflow with approval checkpoints and rules for assigning work. Each output links back to its inputs and the decisions behind it.
OUTCOME
A Python CLI and local dashboard to inspect tasks, decisions and generated files. This is my independent open-source project.
02 / EXPERIENCE
Experience
My work in industrial systems and computer vision.
Jun 2024 – Present
Ghent, Belgium
Yazzoom
Computer Vision Project Engineer
Design hardware and software architectures for industrial computer vision using open-source tools. Lead data acquisition, labeling, model training and deployment across concurrent projects.
Sep 2019 – Jun 2024
Genk, Belgium
Aperam Genk
Computer Vision Engineer
Developed open-source defect-detection and quality-assessment algorithms. Built a Python algorithm using OpenCV and PIL to identify millimetric surface defects in high-resolution coil imagery, rolled out across multiple plants.
Jun 2017 – Sep 2019
Genk, Belgium
Aperam Genk
Reliability Engineer
Initiated the company's first predictive-maintenance data-science project. Installed and monitored more than 1,000 sensors on more than 100 assets, and helped establish a central monitoring room.
Jun 2014 – Jun 2017
Brazil
Aperam South America
Automation Analyst & Telecommunications Engineer
Deployed a trunked radio system serving more than 1,000 employees. Installed and maintained more than 500 security cameras, and maintained routers and servers.
EARLY FOUNDATION
approx. Dec 2013 – May 2014
Developer
RCS Informática • Brazil
Sep 2013 – Nov 2013
Intern
Aperam South America • Brazil
Feb 2013 – Aug 2013
Numerical Calculation Monitor
Universidade Federal de São João del-Rei (UFSJ) • Brazil
03 / SKILLS
Skills
Skills and tools I use at work and in my own projects.
