Brahim Redouane Mellah
Senior Software Developer — machine vision and AI
Montréal, Québec
Experience
2024 — Present
Senior Software Developer Zebra Technologies

Owning algorithms other people ship on
One of the engineers responsible for the library’s algorithms. The work runs past the code: agreeing behaviour with the UI, documentation and test teams before it ships, and going into the applications customers cannot get working.

Inspection that runs on the camera
Moving the imaging library onto the camera itself, so inspection runs on the device rather than on a PC wired to it.
- Built
- Aurora Imaging Library · Aurora Focus smart camera
- Stack
- C++ · machine vision algorithms · embedded / edge imaging
- Tools
- Visual Studio · Git / GitHub · JIRA
- Learned
- A library is a product only when the algorithm, the documentation and the tests agree — and the disagreements surface in a customer’s application long before they surface in a test suite.
2022 — 2024
Software Developer II Zebra Technologies
The geometry underneath the 2D modules
Model Finder, Edge Finder, Measurement, Metrology, Calibration and Bead — solving, implementing and debugging the geometry, algebra and non-linear optimisation underneath them, and widening the tests that guard them. Took the Advanced Geometric Matcher from the algorithm it had been to a public API, designed with the team.

Bringing machine learning into a geometric library
Brought classical machine learning into modules that had been purely geometric, and documented the new APIs and customer examples that went out with them. Supervised trainees and interns.
- Built
- Advanced Geometric Matcher API · MIL 2D modules · customer-facing API examples
- Stack
- C++ · geometric matching · metrology · non-linear optimisation · classical ML
- Tools
- Visual Studio · Git / GitHub · JIRA · Agile
- Learned
- Writing the documentation and the examples is where an API design gets tested — the questions that come back are about the interface, almost never about the algorithm.
2019 — 2022
Software Developer Matrox Imaging
Matrox Imaging became part of Zebra Technologies in 2022; the work continued without a break.

Learning the industry
First job after the master’s, on the modules that measure things: Calibration, Measurement and Metrology. Designed the Advanced Geometric Matcher, the algorithm that later became a public API.

Working inside a large C++ codebase
Replaced older solvers in the geometry modules with non-linear optimisers, Levenberg–Marquardt among them, in a library whose existing behaviour thousands of installed applications depend on.
- Built
- Pattern matching · Edge detection · Calibration · Measurement and metrology
- Stack
- C++ · non-linear optimisation · Levenberg–Marquardt · classical ML
- Tools
- Visual Studio · Git · SVN
- Learned
- Writing a better algorithm is the short half of the job; getting it into a library other people already depend on, without changing an answer they rely on, is the long one.
2017 — 2019
M.Sc., Control Systems Engineering Polytechnique Montréal
GPA 3.87/4 · Al Ghurair Foundation Scholarship — 1 of 100 selected from more than 15,000 applicants.

How I got to Canada
The scholarship was awarded on academic merit and paid for the master’s. It is the reason the rest of this page happens in Montréal.

Systems, not only coursework
Every course that mattered ended in something that had to run: a robot that had to reach the point, a controller that had to hold its deadline, a camera that had to tell the arm where the part was.
- Built
- Control theory · Robotics · Image processing
- Stack
- C++ · digital control · detection and estimation · stochastic and robust control · image processing
- Tools
- QNX · ROS · MATLAB / Simulink · Fanuc robots
- Learned
- A control law that is correct on paper and late by ten milliseconds is wrong — the deadline is part of the specification, not a detail of the implementation.
2012 — 2017
B.Eng., Electrical Engineering — Control Systems École Nationale Polytechnique, Algiers
GPA 17.5/20 · graduated 5th of 1,400 — top 1% nationally.

Two years of preparatory classes first
Mathematics, physics and programming, nine to six, six days a week, with the ranking at the end deciding which school you enter. It is where the habit of working a problem until it is actually solved came from.

Engineering that left the classroom
The degree ended in control systems built against real hardware and real factory constraints rather than simulations of them.
- Built
- Adaptive cruise control for an autonomous vehicle · PLC program for an industrial assembly machine
- Stack
- C · MATLAB · VHDL · control design · process identification · optimal control
- Tools
- Unity-Pro (Schneider) · Simatic-Manager (Siemens) · Simulink · LabVIEW
- Learned
- Hardware does not care how elegant the model is: the first assembly machine I programmed spent its faults on sensors and wiring, not on the control logic I had spent the term on.
Skills
- Languages
- C++ C Python Dart / Flutter TypeScript MATLAB VHDL SQL
- Machine vision
- Aurora Imaging Library Matrox Imaging Library OpenCV geometric matching calibration metrology edge measurement
- Maths & algorithms
- non-linear optimisation Levenberg–Marquardt control theory classical ML deep learning multi-agent pipelines
- Cloud & backend
- Google Cloud Firebase Cloud Functions Firestore Cloud Storage FastAPI Prefect
- Systems & tools
- real-time systems QNX ROS Git / GitHub SVN Simulink LabVIEW Unity-Pro (Schneider) Simatic-Manager (Siemens) JIRA
- Spoken
- French English Arabic
Certifications
- 2023 Neural Networks and Deep Learning · Convolutional Neural Networks DeepLearning.AI
- 2023 Machine Learning: Classification University of Washington