Software engineering for digital forensics.
YCAY
[2025]
YC
「يوسفشواي」
AY
YOUSSEF [CHOUAY] ↓
software engineer, AI researcher and graph theory enthusiast
Scroll
YCAY[HOME]IS THE FOLIO OF
YOUSSEF[يوسف شواي]CHOUAY ↓
SOFTWARE ENGINEER, RESEARCHER
andGRAPH THEORY ENTHUSIAST
AI RESEARCHER @ NATIONAL RESEARCH COUNCIL
BASED IN OTTAWA, CANADA
I'm a software engineer and researcher with a passion for creating elegant, efficient solutions to complex problems. Growing up curious about the world and my surroundings, I've always wanted to understand how they worked.
M.SC COMPUTER SCIENCE @ UOTTAWA
FOCUS: GRAPH THEORY
I've kept that approach and mindset throughout my whole life and I'm slowly chipping the iceberg of knowledge hoping to make a big enough dent to make my younger self proud. Currently pursuing a Master's in Computer Science with a focus on Graph Theory, and working as a software engineer at Magnet Forensics. I'm constantly exploring new technologies and ways to contribute.
ED
「تعليم」
UInstitution
University of Ottawa
Ottawa, ON / 2020 — 2026
20202021202220232024202520262027
Now
In progress · 88%Jan 2025 — Dec 2026
Master of
Computer Science
Thesis
Minimum Feature Size in Proper Triangulations: A tight logarithmic bound
- Supervisor
- Vida Dujmović
- Funding
- $52,000+ in scholarships
EnrolledExpected · Dec 2026
EX
「خبرة」
P01 MagnetNov 2026 – Present
'22'23'24'25'26'27
4×
faster inferenceLLM threat detection on the AI Security team.
- Worked with the AI Security team to migrate a prompt-scan engine to state-of-the-art detection models for LLM threat analysis, including implementation, testing, debugging, and rollout support.
- Implemented chunking behavior for an inference engine in Rust and Go, improving support for long prompts and document-style inputs in AI security workflows.
- Introduced dynamic chunking to the inference engine, speeding up inference by 4× and cutting energy usage by 62% while raising throughput.
- Owned design, implementation, testing, and AWS deployment of a backend service that retrieves company profiles and market data through MCP-based integrations for cybersecurity research workflows.
- Rust
- Go
- AWS
- MCP
- LLM security
56%
lower maintenance costsFirst-author research putting LLM agents inside buildings and utilities.
A
AI-Enhanced Building Automation (BAS)
- Published, as first author, the peer-reviewed paper “Implementing AI in Smart Buildings: A Modular, Proof-of-Concept approach” and presented it at IEEE EPEC 2025.
- Designed and deployed Python/LangChain agents bridging BAS and LLM tools, cutting data-processing time and operator workload by 49%.
- Built a SQLite-backed ingestion pipeline to process and integrate real-time BAS streams reliably.
- Partnered with Delta Controls and Carleton University to deliver AI building agents, achieving a 56% reduction in maintenance costs via automated fault detection, predictive maintenance, and real-time alerts.
B
Enhanced Utilities Chatbot
- Built a multi-agent, tool-using utilities chatbot that explains bills, simulates alternate rate plans, and diagnoses anomalies by linking AMI data with weather/holidays/tariffs for a $3M+ annual revenue company.
- Implemented a time-series engine (seasonal decomposition + change-point detection) to flag spikes, persistent baseload, and overnight leaks, then auto-generate plain-English "why it happened" narratives and savings playbooks.
- Python
- LangChain
- Multi-agent
- SQLite
- Time-series
10+
classes taughtGraduate ML and undergraduate algorithms courses.
- Assisted in teaching graduate classes (Machine Learning for Bio-informatics) and undergraduate courses (Data Structures & Algorithms, Design & Analysis of Algorithms, Programming Paradigms).
- Machine learning
- Bioinformatics
- Algorithms
- Paradigms
90%
faster queries & UIAn automation dashboard, end to end.
- Designed and implemented an Automation Dashboard using Angular, TypeScript, and Django, with a PostgreSQL database.
- Achieved over 90% faster query execution and UI responsiveness by optimizing API endpoints and reducing frontend rendering times.
- Angular
- TypeScript
- Django
- PostgreSQL
80%
faster search responsesRebuilt the university's search engine.
- Redesigned and optimized the university's search engine using PHP, MySQL, and Apache, improving query response times by 80%.
- Developed automation scripts using PHP, Bash, and Cron jobs to enhance search speed by 54%.
- PHP
- MySQL
- Apache
- Bash
- Cron
QUIETLYpowerful
DIGITAL EXPERIENCESEngineering is a bridge between logic and creativity. My goal is to build solutions that are not just functional, but elegant, quietly threading innovation into every line of code.
View All Projects
Coming Soon
Selected Works
Fueled by curiosity, shaped by codeYC © 2026
BLOG
Writings & research / Field notes
A few things I’ve built, questioned, and tried to make sense of. Pull on a thread.
When markets misbehave.
A stochastic processes project about one dangerous sentence: these two risks are independent.
Open the note · Nov 30, 2025
Can a language model play chess?
I built a benchmark to find out, then rebuilt it twice. The hardest part was never the chess.
Open the note · Jul 10, 2025「موسيقى & سفر」
Recent Travels
memories
Currently On Repeat
RilesSometimes45 RPM
Riles - SOMETIMES
Ask Youssef's AI anything