I'm Fayaz, a Computer Engineering student at York University passionate about building full-stack apps, AI-powered tools, and creative software experiences. I enjoy working across web, cloud, data pipelines, and automation, and I love turning ambitious ideas into polished products.

I've worked on internal enterprise software, LLM-powered apps, social platforms, hackathons, and game projects. I’m especially interested in software engineering, AI systems, developer tools, and building technology that helps people.

TD Securities – Research and Compliance, Corporate & Investment Banking
Jan 2026 – Present

• Built a full-stack internal web application using Java, Spring Boot, Python, and React to streamline audit, risk, and compliance workflows.
• Helped create a unified platform for Technology Asset Owners across Corporate and Investment Banking to centrally view and manage audit records.
• Designed cloud infrastructure using Azure SQL, Azure Redis Cache, and Azure Sentinel for scalable and secure enterprise operations.
• Developed an Azure batch job to automate manual data scrubbing from Archer and Jira, improving data quality and reducing repetitive work.
• Engineered an ETL pipeline with Python, Selenium, and FastAPI that integrated directly with the app and removed manual upload steps.
• Created a Power Apps Process Improvement Intake Form to improve visibility into internal feedback and operational pain points.

Outlier
2024 – 2025

• Improved LLM output quality by designing, testing, and refining prompts for advanced reasoning and math-intensive tasks.
• Validated AI-generated solutions using Python and MATLAB to strengthen evaluation quality and improve response selection.
• Ranked candidate outputs, analyzed error patterns, and identified failure modes to improve prompt robustness and model reliability.
• Built strong intuition around prompt engineering, model behavior, and evaluation workflows for production-style AI systems.

Radar
May 2022 – Aug 2022

• Contributed to a social matching app using Flutter on the frontend with PostgreSQL and Go for backend support.
• Designed and implemented core user features such as match, un-match, block, and unblock.
• Improved the beta product’s user experience with graphical UI updates and app functionality enhancements.
• Managed user data and SQL workflows while using Docker to support development and deployment processes.

• Spring Boot
• FastAPI
• Selenium
• Puppeteer
• Azure SQL
• Azure Redis Cache
• Azure Sentinel
• Power Apps
• SharePoint
• CI/CD
• REST APIs
• ETL Pipelines
• Excel

Next.js • TypeScript • LiteLLM • Three.js • Tailwind CSS

• Built a multi-model AI platform where language models compete in a real-time 3D chess environment.
• Implemented server-side move validation, structured JSON parsing, and fallback handling for malformed outputs.
• Designed configurable model matchups, provider routing, move logs, and observability tooling for testing LLM decision-making.

Next.js • TypeScript • Supabase • Puppeteer • Groq API

• Built an AI grocery planning app that combines meal recommendation logic with nutritional and grocery data.
• Integrated LLM-powered workflows to generate personalized, budget-aware meal suggestions.
• Added authentication, rate limiting, serverless APIs, and error handling to make the product deployment-ready.

C • GBDK • GB Studio

• Developing a Game Boy game with custom gameplay systems and original retro-style presentation.
• Successfully got the game running on real Game Boy hardware and a custom cartridge.

Next.js • Supabase • Tailwind CSS • Auth

• Designing a full-stack team management tool for storing, organizing, and viewing Pokémon Showdown teams by metagame.
• Focused on clean UI, persistent storage, and easy team reuse.

• Full-Stack Development
• AI Engineering
• LLM Applications
• Cloud & Infrastructure
• Data Pipelines & Automation
• Developer Tools
• Open Source
• Creative Coding & Game Development

I’m always down to connect, collaborate, or chat about tech, projects, hackathons, and creative ideas.