Ali W. Bekheet,
Engineering Physics & Computing
Some things about me:
- Engineering physics and computing graduate from Queen's University who enjoys building systems meant to last.
- Spent time volunteering in education and student support; teaching and mentoring shaped how I think about explaining technical ideas and designing for real users.
- I like problems where getting something to work is the easy part, and making it reliable, observable, and maintainable is the real work.
- I prefer compounding projects over time rather than one-off experiments.
Some things I'm interested in:
- System design and architecture: turning complex systems into ones easier to reason about and debug.
- Hardware-adjacent problems like sensors and perception.
- Teaching, documentation, and knowledge sharing.
- Learning from well-built systems and good technical writing.
Contact me @
Library Sections
Technical Skills
Programming Languages
- Python
- C++
- Bash / Shell
- SQL
Machine Learning & AI
- PyTorch
- HuggingFace Transformers
- NVIDIA NeMo
- LLM APIs (OpenAI, Azure)
LLM Orchestration
- LangChain
- LangGraph
- Pydantic
- Structured Generation
Backend & Data
- FastAPI
- MongoDB
- Elasticsearch
- Web Scraping
Infrastructure & MLOps
- Docker / Compose
- Kubernetes
- NVIDIA NIM
- CI/CD Pipelines
Cloud Platforms
- Azure / Azure OpenAI
- AWS
Engineering Practices
- System Design
- Observability
- Testing
- Optimization
Languages
- English (Native)
- Arabic (Native)
Recent Writing
Thoughts on AI, engineering, and building things that work.
AI / Reinforcement Learning
I Trained a Bot to Play Catan. It Barely Learned Anything.
Why deep reinforcement learning struggles with board games that aren't Go or Chess—and what I learned from a 28% win rate.
Newsletter
More on Substack
Explore all posts on engineering, AI systems, and lessons learned from building things.