Liron Katsif

Cooking at UWaterloo

Building cool stuff

Me

I'm a Software Engineering student at the University of Waterloo interested in AI, quantitative systems, and building sick software. I enjoy working on machine learning models for financial markets, real time systems, and strong backend architectures.

Outside of tech, I'd like to say that I'm an outdoorsman, and I love spending time in nature, whether that be exploring new places or just disconnecting from a screen. I'm motivated by hard problems, learning fast, and building things that people really use.

Languages

Python, C++, Go, Java, C#, HTML/CSS, C, Swift, JavaScript, TypeScript

Frameworks

PyTorch, React, TensorFlow, MongoDB, REST API, GraphQL, MySQL, OCR

Developer Tools

Git, PyCharm, VS Code, Jira, Postman, Visual Studio, GitHub, Xcode

Education

Bachelor of Software Engineering

Sep 2025 – Apr 2030

University of Waterloo

Honours/Co-op

Experience

Full Stack Engineer

Aug 2025 – Present

Savi Finance • Toronto, ON

  • Developed features used by 1000+ users for the app with React, integrating UI with backend services using Go
  • Architected and developed Social Circles system, including invitation workflows and relationship models
  • Implemented real-time purchase visibility, enabling 1000+ connected users to view each other's activity seamlessly
  • Built and optimized 30+ GraphQL APIs for secure, real-time financial and transactions data
  • Achieved 30% improvement in website SEO by reducing load times with asset optimization and caching

Machine Learning Engineer

Dec 2025 – Present

Wat Street • Waterloo, ON

  • Developed a Volatility Contagion Explorer to model how market volatility propagates across 1000+ assets
  • Represented assets as nodes in a dynamic graph, capturing dependencies and 20000+ temporal relationships
  • Reduced forecasting error by 30% versus GARCH and LSTM models by capturing structural market correlations
  • Engineered the model to learn inverse relationships, identifying assets likely to appreciate during volatility shocks
  • Built robust data pipelines to construct and update evolving market graphs from time varying financial data

Programming Tutor

Aug 2023 – Aug 2025

Self Employed • Toronto, ON

  • Mentored 30+ students in algorithms, data structures, and OOP, boosting programming proficiency
  • Taught best practices across Python, C#, Java, and C, improving code quality and problem-solving skills
  • Delivered 500+ hours of personalized one-on-one tutoring, adapting teaching strategies to individual learning
  • Improved average student grade by 10% by identifying learning gaps and reinforcing core concepts

Mathematics Tutor

Jan 2022 – June 2025

Self Employed • Toronto, ON

  • Provided individual and group tutoring to students in grades 4–11, improving their understanding of concepts
  • Prepared students for assessments and competitions, increased average mark around 10%
  • Developed customized lesson plans and extra curriculum content to enhance learning outcomes
  • Received 30+ positive feedback from clients for clear communication and effective teaching methods

Projects

Chess Engine (Chess-Hacks)

Python PyTorch NumPy

Built a 1700 ELO chess engine using alpha-beta pruning with iterative deepening and transposition tables. Integrated a NNUE evaluation network in 24 hours to provide accurate board scoring with minimal overhead. Achieved high performance by combining NNUE evaluation with optimized pruning, enabling a depth of 6-7. Designed engine architecture separating search, evaluation, and game-state to improve scalability and extension.

Recipe Management

C# .NET REST API

Designed and developed a full-stack application for recipe and meal planning management. Integrated Spoonacular API to fetch thousands of recipes and ingredients, enabling users to search, sort, and save. Implemented features for meal planning, nutritional tracking, and custom recipe creation.

Page Position Extension

JavaScript HTML CSS

Built a Chrome extension that saves and restores scroll positions on webpages to prevent losing your place. Designed a clean, user-friendly popup UI for naming, saving, jumping to, and deleting bookmarks. Published on the Chrome Web Store with cross-tab functionality and persistent storage.

Wrestling Team Manager

C# .NET

Created a full-stack application to help wrestling coaches manage athlete data and performance metrics. Enabled coaches to input, update, and analyze wrestler statistics, improving team performance tracking. Successfully adopted by a high school wrestling coach for team management.

UW Flow Course Ranking

Python Selenium Web Scraping OCR

Developed a Python tool to rank courses by increasing difficulty using data from UWFlow.com. Enabled input via text lists or course list images through OCR integration, improving accessibility. Utilized Selenium for web scraping to retrieve course difficulty ratings automatically.

MacAppSwitcher

Swift macOS AppKit

Built a native macOS application that replicates Windows-style Alt+Tab functionality for window-level switching. Implemented live thumbnail previews with MRU ordering using CGEventTap for hotkey handling and Accessibility APIs. Enables users to switch between individual windows rather than just applications using Cmd+~ shortcut.

Spotify Swipe Queue

JavaScript Chrome Extension CSS

Created a Chrome extension adding mobile-like swipe gestures to Spotify Web for intuitive song queuing. Implemented CMD/Ctrl + drag right functionality with smooth animations and visual feedback. Works across playlists, search results, albums, and artist pages using MutationObserver for dynamic content.

Spotify Queue - Raycast

TypeScript Raycast API OAuth

Developed a Raycast extension enabling users to search for songs and add them directly to their Spotify queue. Integrated Spotify Web API with OAuth authentication for seamless queue management. Streamlines music discovery without switching applications for Raycast power users.

Hand Digit Recognition

Python Machine Learning Computer Vision

Implemented machine learning models for hand-written digit recognition and classification. Explored various ML algorithms and neural network architectures for image classification tasks. Built using Python with focus on computer vision and pattern recognition techniques.

Get In Touch