Three applied courses connect machine-learning foundations to systems people actually use. Each course combines concepts, implementation, and critical evaluation—moving from an idea on the whiteboard to a working system that can be measured and improved.
Information Retrieval
Build the machinery behind search—from inverted indexes and ranking to vector databases, RAG, and agent memory.
Explore the course → CSCI 4146 / 6409Process of Data Science
Frame decisions, audit model behavior, diagnose AI systems, reason causally, and monitor production outcomes.
Explore the course → CSCI 6517Recommender Systems
Progress from collaborative filtering to sequential, conversational, and large-scale multi-stage recommenders.
Explore the course →Understand it. Build it. Test what breaks.
Students learn through a recurring loop: establish the mathematical idea, implement it on realistic data, choose metrics that reflect the real goal, and inspect the failures. The emphasis is not only on producing a model, but on understanding the system around it.
Previously taught: MIE 451/1513, Decision Support Systems, University of Toronto, Winter 2020.