Mithun

Sivapathasundram

About

Acomputersciencestudentwithapassionforbuildingclean,performantsoftware.

Interestedinfull-stackdevelopment,machine learning,andtheintersectionofgreatengineeringandgreatdesign.

Based in

Toronto, ON

Education

BSc Computer Science @

Toronto Metropolitan University

4th Year · 2026

Looking for

2026 New Grad Roles

Get in touch

Experience

Software Engineer Intern

May 2025 – Dec 2025

Environment and Climate Change Canada · Toronto, ON

Shipped features and improved data integrity on a national reporting platform serving 12,000+ users.

Delivered 15+ features (facility transfer workflows, bulk import tools, automated notifications) on a C#/.NET 8 platform serving 12,000+ users across Canada
Diagnosed and resolved 20+ production defects across front-end and C#/.NET server-side code, raising automated test coverage from 65% to 85% within a Git branch-and-PR workflow
Validated and normalized high-volume reporting data across 10+ interfaces using C#/.NET validation rules and MS SQL checks, cutting invalid data by 30% and improving the reliability of structured data surfaced to users
Wrote MS SQL stored procedures that identified and cleaned 4,000+ corrupted records, improving data integrity by 95% and saving 18+ hrs/week of manual validation

C#.NET 8Entity FrameworkMS SQLAzure DevOpsXML

Software Engineer Intern

Jan 2024 – Sep 2024

Ontario Treasury Board Secretariat · Toronto, ON

Built full-stack features and automation tools for legacy Ministry applications used by 500+ employees.

Shipped 20+ full-stack features across 2 legacy web apps, adding custom JS validation and REST API integrations that standardised data entry and cut downstream processing errors
Developed 25+ responsive, reusable front-end components from Figma specifications using custom CSS, partnering with UI/UX designers to translate wireframes into production UI and reducing design-to-dev handoff time by 30%
Designed relational schemas and optimized SQL queries powering an automated ETL data pipeline that delivered same-day data to downstream consumers, eliminating 8 hours/week of manual work
Built a full-stack tool with Power Automate's AI Builder that flagged AI-written applications at 95% accuracy, earning 2nd of 30+ teams at an internal hackathon
Built and documented REST APIs over JSON with layered automated testing reaching 90% coverage, gating release quality in a CI/CD pipeline before merge

JavaScriptHTML & CSSSQLPower BIREST APIsAzure DevOps

Projects

2026

Fantasy Basketball Companion

Full-stack sports analytics platform with a real-time AI coaching chatbot.

Next.jsTypeScriptSupabaseGoogle Gemini APIVitestPlaywrightReact Testing LibraryRedisLangChainDockerKubernetesGKE

Fantasy Basketball Companion

GitHub Live

→Shipped a full-stack fantasy basketball platform with live stats, matchup analysis, and AI-powered roster recommendations, deployed on Vercel and serving real-time analytics to active users
→Integrated LangChain.js into the AI coaching engine with ConversationBufferMemory, enabling persistent conversational context across roster recommendations with low-latency streaming
→Implemented Redis-backed distributed rate limiter using Upstash, enforcing a 3 req/min cap across concurrent Next.js serverless instances
→Decomposed a monolithic Next.js/Supabase app into three independently deployable services using Docker multi-stage builds, Kubernetes, and GKE Autopilot, configuring a HorizontalPodAutoscaler on the AI inference service to scale 1 to 5 replicas on CPU load
→92% statement coverage across unit, component, and E2E test suites using Vitest, React Testing Library, and Playwright

2025–2026

NBA Predictive Analytics Platform

End-to-end ML pipeline predicting per-game NBA player stats at 75%+ accuracy.

Pythonscikit-learnFlaskpytestGitHub Actions

NBA Predictive Analytics Platform

GitHub Live

→6 ensemble models covering 500+ active players with automated daily retraining on fresh season data
→GitHub Actions ETL pipeline scraping and normalising 10,000+ player stats daily from Basketball Reference with zero manual intervention
→MLOps infrastructure on Hugging Face Spaces (16 GB RAM) with CI/CD pipelines serving real-time predictions 24/7

2026

SimpleBL

AI fitness research assistant that grounds every response in live peer-reviewed PubMed literature.

ReactTypeScriptNode.jsExpressGroq APIPubMed / NCBITailwind CSSVite

SimpleBL

GitHub Live

→Built a full-stack research tool that fetches live PubMed abstracts via NCBI E-utilities on every query with no vector DB or cached data so responses always reflect current literature.
→Integrated Groq API (llama-3.3-70b-versatile) to produce structured JSON search results and markdown chat replies, with inline citation pills linking every claim to its primary source.
→Implemented a 340ms inter-request rate limiter to comply with NCBI's 3 req/s policy, with graceful degradation ensuring PubMed failures never surface as unhandled errors.

2025

VetConnect

School Project: Full-stack veterinary management system connecting pet owners and clinics.

TypeScriptReactSupabasePostgreSQLHTML & CSS

VetConnect

→Collaborated with 4 developers to design and ship a secure, real-time patient management platform handling appointments, records, and messaging.
→Implemented role-based access control via Supabase Row-Level Security, ensuring strict data privacy between pet owners and veterinary staff.

2024

Smart Recruiter

AI-powered recruitment platform. Placed 2nd at the OPS Phenomenal Hackathon.

Microsoft Power AppsJavaScriptREST APIsMachine Learning

Smart Recruiter

→Reduced manual screening time by integrating Microsoft's Category Classification Model to flag AI-generated job applications.
→Delivered a full end-to-end platform within a tight hackathon timeline using Microsoft Power Apps and REST APIs.

2024

Olympic Medal Predictor

Linear regression model forecasting Olympic medal counts from 100+ years of historical data.

Pythonpandasscikit-learnseabornJupyter Notebook

Olympic Medal Predictor

GitHub

→Trained and evaluated the model across multiple Olympic Games, achieving strong predictive accuracy for country-level medal outcomes.
→Built end-to-end data cleaning, feature engineering, and interactive visualisations using pandas, seaborn, and scikit-learn.

2024

Soccer League Database System

School Project: Relational database system managing all data for a competitive soccer league.

SQLOracle DBJava

Soccer League Database System

→Designed a normalised Oracle DB schema covering teams, players, fixtures, results, and standings with full referential integrity.
→Conducted end-to-end testing to verify data consistency and reliability across concurrent read/write operations.

2023

Java Stock Market System

School Project: OOP stock market simulator with real-time data, built by a team of 6.

JavaJUnitAlpha Vantage API

Java Stock Market System

→Engineered comprehensive UML models to drive a clean, extensible OOP architecture across a 6-person team.
→Integrated the Alpha Vantage API for live stock data and achieved full unit test coverage with JUnit.

Skills

Languages

PythonJavaTypeScriptJavaScriptC#SQLBashHTMLCSS

Frameworks/Libraries

React.jsNext.js.NET 8Entity FrameworkFlaskTailwind CSSpandasscikit-learnRequestsSeleniumRedisLangChain

Databases & Cloud

PostgreSQLMS SQLMongoDBSupabaseDockerKubernetesGKEAzureHugging Face SpacesVercel

Tools & Workflows

GitGitHub ActionsFigmaCI/CDTest-Driven DevelopmentAgile MethodologiesREST APIsMLOpsPowerBIOutSystemsClaude CodeGitHub Copilot

Testing

VitestReact Testing LibraryPlaywrightpytest

Let's work together!

Full-stack developer | Interested in ML | 2026 new grad | Passionate about building software that makes an impact.

Version

2026 © Edition

Timezone

EST (GMT‑5)

Socials

GitHubLinkedIn