Candidate for Bachelors of Mathematics Co-op student at the University of Waterloo specializing in Applied Mathematics - Scientific Machine Learning and Statistics, with a minor in Computing.

I build machine learning systems for financial services, with an emphasis on rigorous evaluation, interpretability, and production constraints. My work focuses not just on whether models perform well in isolation, but whether they are reliable, understandable, and genuinely useful in real-world settings.

AI Product Engineer Intern — Ibreez Academy Inc. Demo
Built SIRS, a Python + Supabase student-operations platform for 3,000+ students across 27 course types. Engineered 7+ operational workflows across 3 synchronized Streamlit apps spanning registration, payments, kiosk/live attendance, exam tracking, ReportLab PDFs, and real-time analytics for lead generation and revenue.

ML Engineer — Wat Street (University of Waterloo) Demo
Collaborated on a GAT + LSTM system forecasting next-session volatility for 155 of 500 S&P 500 equities. Implemented the GATLayer (10 features → 32D embeddings, message passing, graph-to-LSTM interface) and built the interactive product tour for exploring forecasts, network influence, risk rankings, and shock propagation.

AI/ML Engineer Intern — Paideia Mundi NeuroSystems Institute (NDA-Protected)
Contributed to a confidential multimodal ML research prototype supporting children's neurodevelopmental health. Engineered EEG connectivity features and a custom scikit-learn compatible feature selector with non-parametric ranking and multi-target selection logic, validated through a full pytest test suite.

AI Engineer — Dazia Consulting Inc. Demo
Expanded a RAG pipeline by adding semantic chunking, indexing 1500+ pages of CSC/IFIC content across 1,600 embeddings per dataset in ChromaDB, and deploying the vector store with Docker for persistent storage. Built 8 Express.js REST APIs with CORS and rate limiting, scaling backend across 4 AI tutoring services. Implemented a 20-interaction per-service freemium cap aligned with cost modeling and reduced Gemini 2.5 Flash API inference costs by 30% through retrieval optimization, integrating the backend with a React frontend.

Co-Founder — Wanderers Website
Co-founded Wanderers, a live campus social platform enabling University of Waterloo students to form real-life connections through spontaneous activity bubbles, selected and pitched at Velocity Pitch Competition 2026. Engineered the backend integration layer wiring Next.js API routes to Supabase for live feed updates and a Gemini 2.5 Flash pipeline for converting natural language into structured bubble creation in under 2 seconds.

Languages:

Machine Learning:

Data & Infrastructure:

Web & Visualization:

• LinkedIn: https://www.linkedin.com/in/kapiliyer29
• Email: k22iyer@uwaterloo.ca
• Portfolio: https://kapil-iyer-portfolio.vercel.app
• Resume: https://drive.google.com/file/d/1Uu-O4c6otsU8rn-SuORZ4_Y9NK7Oif6_/view?usp=sharing