Thea Ishida
Software Engineering · Fall 2026

Thea
Ishida.

CS student at Western University. Varsity athlete. Builder of systems that matter — from ML infrastructure at Autodesk to real-time financial engines. Drawn to complexity, especially in the world of finance.

View My Work → Get In Touch
Scroll
About me

Engineer.
Athlete.
Builder.

I'm a 4th-year Computer Science student at Western University, where I'm also the Team Captain of the Women's Varsity Volleyball team — a role that's sharpened my instincts for leadership under pressure as much as any codebase has.

I'm drawn to building solutions to genuinely hard problems, with a particular pull toward the financial space — where systems have to be precise, fast, and reliable all at once. Whether that's a real-time arbitrage engine or a high-throughput banking pipeline, that intersection of complexity and consequence is where I do my best work.

Currently interning at Autodesk in Toronto, building AI-powered infrastructure that replaces legacy systems at scale. Previously at RBC Capital Markets and General Dynamics.

3.8
GPA · Dean's Honour Roll
3×
Internships completed
2×
Hackathon 1st place
2yr
Varsity team captain
Python Java TypeScript C# SQL MATLAB Docker Kubernetes AWS Lambda Spring Boot React Flask Kafka WebSockets REST APIs Git / CI/CD MCP Azure (AZ-900)
Work history

Where I've
built things.

Autodesk
May 2026 – Present
Toronto, ON
Cloud Platforms &
Product Access
Software Engineer Intern
Autodesk ↗ is a global design and engineering software company behind tools like AutoCAD, Revit, and Fusion. The Cloud Platforms & Product Access team owns the infrastructure and services that power product access, licensing, and identity across Autodesk's platform. As a software engineering intern, I built an AI-powered Slack agent to replace the team's legacy support bot, and re-architected the RAG pipeline powering it — cutting incident triage from hours to seconds and expanding context capacity by 16×.
  • Building an AI-powered Slack Agent platform using Python, Bold SDK, AWS Lambda, MCP, Kubernetes, and Docker — designed to replace a legacy support bot and cut incident triage time from hours to seconds.
  • Re-architected the internal agent's RAG pipeline with a company-approved embedding model and a CloudOS-managed vector database, achieving 16× context capacity and 4× dimensionality with zero added infrastructure costs.
Python AWS Lambda MCP Kubernetes Docker RAG LLMs
Visit Autodesk →
General Dynamics
Jan 2026 – May 2026
London, ON
Vehicle Software
Applications
Software Engineer Intern
The Logistics Vehicle Modernization (LVM) Project ↗ is a Canadian Department of National Defence contract to modernize the military's fleet of logistics vehicles. On the Vehicle Software Applications team, I engineered a bilingual (English/French) internal portal providing operators with full control over CANbus devices and real-time telemetry from complex vehicle systems — bridging a C#/.NET data layer with a modern Vue.js + Quasar frontend and scaling localization to 100% of the user base.
  • Developed a responsive Vue.js + Quasar (TypeScript) web dashboard connected to a CANbus component database for real-time low-level system monitoring and diagnostics, with full French/English localization.
  • Engineered a custom C#/.NET data streaming client in a Linux environment, eliminating parsing failures and achieving 100% reliability for critical system data loggers.
Vue.js TypeScript Quasar C#/.NET Linux CANbus
View LVM Project →
RBC Capital Markets
May 2025 – Dec 2025
Toronto, ON
U.S. Cash Management
Payments API Team
Backend Software Engineering Intern
RBC Clear ↗ is RBC Capital Markets' institutional clearing and custody platform, enabling high-volume U.S. payment flows and cash management operations for financial institutions. As a backend intern on the Payments API team, I built core pipeline infrastructure that directly underpins the platform's payment processing capabilities — accelerating delivery cycles and improving deployment reliability through workflow automation and containerized microservices.
  • Accelerated software delivery and deployment cycles by 60% by engineering a high-throughput Java and Spring Boot backend pipeline, leveraging Camunda to automate core workflows.
  • Containerized microservices with Docker and orchestrated via Kubernetes, improving environment consistency and reducing deployment lead times across the team.
Java Spring Boot Docker Kubernetes Camunda
View RBC Clear →
Selected work

Things I've
shipped.

01 · Featured
🟢 In Progress

Real-Time Arbitrage Engine

A high-performance cross-exchange arbitrage calculator that identifies price discrepancies in real time. Uses asynchronous WebSockets to process market signals with minimal latency, and vectorized NumPy calculations to evaluate triangular arbitrage opportunities across multiple currency pairs simultaneously.

PythonNumPyPandasWebSocketsAsync I/O
02
🏆 1st Place · Toronto Tech Expo

SnowFall — IoT Health System

Led a 5-engineer team to build a wearable IoT health system. Synchronized sensors with a Python backend for 100% real-time data fidelity. Built a caretaker-facing chatbot using a 1.3B DeepSeek LLM, improving interaction efficiency by 35%. Won "Best Use of AI for Good."

PythonIoTDeepSeek LLMWearables
03
🏆 1st Place · Western Dev Society

SmartCart — Accessible Retail Platform

Led a cross-functional team to build an AI-driven accessible retail platform. Delivered a feature-complete full-stack MVP in a 48-hour sprint, winning first place at the Western Dev Society Hackathon.

ReactFlaskPythonAI/ML
From GitHub
Fetching repositories from GitHub…
View all on GitHub →
Let's connect

Have something
worth building?

I'm always open to interesting conversations — especially at the intersection of software and finance. Don't hesitate to reach out.

thea.ishida@gmail.com →
LinkedIn ↗ GitHub ↗