Hi, I’m
Cabrel FOSSO
- Full-stack developer
- Explainable AI
- Physics teacher
A full-stack developer with a Master’s in artificial intelligence, I build web applications and work on making AI models explainable. I’ve also been teaching physics since 2020.
- teaching since
- 2020
- Master’s in AI, class of 2026
- Top 5
- interactive simulations
- 2
- Python
- PyTorch
- TensorFlow
- Scikit-learn
- Hugging Face Transformers
- Pandas
- NumPy
- SHAP
- Next.js
- React
- Node.js
- TypeScript
- MySQL
- Docker
- Git
- Linux
About
Three roles, one thread: making things understandable
Teach
Certified secondary-school physics teacher (PLEG grade) in Cameroon’s public service since December 2020, at Lycée Bilingue de ZENMEH in Dschang. I teach physics, chemistry, technology and mathematics, and run the school’s science club. As co-founder of the POLYGÉNIE community, I also tutor computer science students from first to third year.
Research
Master’s in Computer Science (AI) from the University of Dschang, with honours (Mention Bien), ranked in the top five of my cohort. My thesis tackles the explainability of Transformer models for time-series forecasting, benchmarked against PatchTST, Informer, Autoformer and TFT.
Build
Full-stack developer working with Next.js, React, Node.js and Python. Since December 2025, as an intern at Mo’ock, I’ve worked across the front end, back end and deployment of web applications.
Services
What I can do for you
Web development
Websites and web apps, from first mock-up to launch, built with Next.js, React, Node.js and Python.
AI and data science
Exploring and visualizing your data, training machine learning models, and explaining their predictions clearly with explainable AI.
Educational tools and simulations
Learning platforms and interactive simulations, designed by a teacher for real students.
Tutoring and training
Tutoring and training in physics, mathematics and computer science, from secondary school to university.
Research
SEATTS: making Transformer forecasts readable
My Master’s thesis introduces SEATTS, a module attached to the output of a Transformer model. It explains the model’s time-series forecasts through four concepts: level, trend, context and calendar. A paper is in preparation.
Level
The baseline value the series moves around.
Trend
The underlying direction, up or down.
Context
How recent observations shape the forecast.
Calendar
Effects tied to the day, the week or the season.
Skills
My toolkit
Languages
- Python
- JavaScript
- TypeScript
- C
Machine learning
- PyTorch
- TensorFlow
- Scikit-learn
- Transformers (Hugging Face)
- Pandas
- NumPy
Interpretability
- SHAP
- attention maps
Web development
- Next.js
- React
- Node.js
- MySQL
Tools
- Git
- Docker
- Linux
- JupyterLab
- LaTeX
Background
From physics to artificial intelligence
Since December 2025
Full-stack developer (intern), Mo’ock
2024–2026
Master’s in Computer Science, AI track, University of Dschang, top five of the cohort
Since 2024–25
Co-founder of the POLYGÉNIE community
2021–2024
Bachelor’s in Computer Science, University of Dschang, ranked first
Since December 2020
Physics teacher (PLEG), Lycée Bilingue de ZENMEH
2015–2020
DIPES I and II in Physics, École normale supérieure de Yaoundé
What brings you here?
I’m offering a PhD position
For: thesis supervisors, doctoral schools, scholarship committees
I’m looking to become a PhD student, and I’m seeking a lab and a supervisor willing to take me on for a PhD in artificial intelligence, ideally in explainable AI or at the interface of AI and physics. My Master’s trained me in research: a thesis on making Transformer models explainable, with a paper in preparation. Teaching physics since 2020 has taught me how to make complex ideas clear.
I have a project
For: schools, teachers, organizations
I design and build educational web platforms and websites with Next.js and Python, from first mock-up to deployment. Tell me what you need and I’ll get back to you with a proposal within 72 hours.
I’m a researcher
For: researchers and faculty
My Master’s thesis introduces SEATTS, a module attached to the output of a Transformer model. It explains the model’s time-series forecasts through four concepts: level, trend, context and calendar. A paper is in preparation.
The other side
Physics teacher
Since 2020, I’ve been teaching physics at Lycée Bilingue de ZENMEH in Dschang. I also build interactive simulations that show what the equations are saying.
Launch a projectile
Set the launch angle and speed, then watch the trajectory. The equations of motion and the range update as you go.
Launch angle (°)45
Initial speed (m/s)10
- Equations of motion
x(t) = 7.07 t
y(t) = −4.90 t2 + 7.07 t
- Range
- 10.2 m
A PhD position, a project, a collaboration?
Write to me, in French or English.