Syed Faizan
Shah
M.Sc. Computer Science student specializing in AI & Machine Learning at JMU Würzburg, based in Munich. I build things from scratch — from neural networks in pure NumPy to IoT intrusion detection systems on Raspberry Pi. Looking for a working student role where I can apply deep learning and computer vision to real-world problems
Why Work with Me
Built From First Principles
I don't just use ML frameworks — I understand what's underneath. My HCI project required building a complete neural network from scratch using only NumPy: PCA, three gradient descent variants, and real-time gesture classification. It scored 110%. I learn by building from the ground up.
Research to Deployment
I take academic papers and turn them into working systems. For my IoT Security Praktikum, I implemented five USENIX-published intrusion detection systems and packaged them for Raspberry Pi deployment. I'm comfortable reading research, writing the code, and shipping it.
Ready to Contribute Now
I'm an M.Sc. student based in Munich with a valid student visa — no sponsorship needed. I'm available 20h/week during semester and 40h during breaks. I've co-founded a business and worked in fast-paced teams, so I don't need hand-holding to get started.
What I've Built
Gesture-Controlled Slideshow & Tetris — HCI Project
(2026)
IoT Intrusion Detection Systems — Security Praktikum
(2026)
MALGANs: GANs for Cybersecurity — Coursework
(2025)
MALGANs: Generative AI for Cybersecurity
2024
1. Read the Research
I start with the literature. For my IoT Security Praktikum, I studied five published USENIX systems end-to-end before writing a single line of code. I believe understanding the "why" behind a method matters as much as the implementation.
2. Build From Scratch
I prefer to understand things at a fundamental level. My HCI project required implementing neural networks, PCA, and gradient descent using only NumPy — no ML libraries allowed. This approach gives me a deeper grasp of the tools I later use in frameworks like PyTorch.
3. Iterate & Test
I write code that actually runs. Every system in my IoT project includes integration tests, handles edge cases, and was validated against the original paper's results. I use Git, Docker, and CI-friendly workflows to keep things reproducible.
4. Document & Deliver
I treat documentation as part of the work, not an afterthought. My Praktikum report is a 9-chapter LaTeX document with architecture diagrams, and I caught 7 factual errors during a final cross-reference pass. Clean handoffs matter.
What I Work With
AI & Machine Learning
I've built ML systems from scratch in academic and personal projects, from neural networks in pure NumPy to GAN architectures for cybersecurity research.
Deep Learning & Neural Networks
Computer Vision (OpenCV, MediaPipe)
Reinforcement Learning (PPO)
Security & Systems
I implement research-grade security systems. My Praktikum covered five published intrusion detection architectures, deployed and tested on real IoT hardware.
IoT Intrusion Detection
Software Exploitation (academic)
Linux, Docker, Raspberry Pi
Data & Tools
I work daily with Python's data stack for analysis, visualization, and building data pipelines for my projects.
Python, NumPy, Pandas, Matplotlib
Git, Jupyter, Docker
SQL (basic), LaTeX
A Bit About Me
I’m Syed Faizan Shah, a Master’s student in Computer Science (AI & Machine Learning) at JMU Würzburg, currently based in Munich. I like building things from the ground up — whether that’s neural networks without frameworks, intrusion detection systems from research papers, or a receipt-scanning app to track my grocery spending in Germany. Before my Master’s, I co-founded a small hardware resale business and worked in e-commerce and data annotation. I’m now focused on deepening my skills in deep learning, computer vision, and security — and looking for a Werkstudent role where I can contribute to real projects.
110%
HCI Project Score
Built a gesture-controlled system with neural networks from scratch using only NumPy — the highest score in the course.
5
Published Systems Implemented
Reproduced five USENIX Security intrusion detection systems for IoT, with full integration tests and Raspberry Pi deployment.
1.3
AI Research Seminar Grade
Top grade in research seminars covering deep learning, computer vision, and generative AI at JMU Würzburg.