Activities
...
Off
2:40 Remaining (75%)
Settings
Lock
Power Off / Log Out
Jiayi Yang — AI Systems & Retrieval Researcher
CS Specialist @ University of Toronto
I'm a Computer Science Specialist (Math Minor, ASIP Co-op) at the University of Toronto, St. George campus, focusing on AI, Theory of Computing and Computer Systems. ( Reach me at admin@yangjiayi.top :) ) I'm a Research Assistant at the Graph Foundry Lab (Prof. Nick Koudas), building TriDex — a searchable long-term memory index for robots. My research is on approximate nearest neighbour search, vector databases, retrieval-augmented generation and agent memory. Write-ups live on my research index. Since 2019 I've been running my own self-hosted servers — websites, mail servers and all the network plumbing in between. I'm open to 4–20 month co-op terms.
Education
- University of Toronto, St. George — Computer Science Specialist & Math Minor - ASIP (Co-op) (Aug 2024 - Jan 2028). Focus: AI, Theory of Computing, Computer Systems · open to 4–20 month co-op terms
- International Baccalaureate Diploma — Chengdu, China (Aug 2021 - May 2024). Math AA HL 7/7 · Physics HL 7/7 · CS SL 6/7 · Math EE B/A
Experience
- Research Assistant — TriDex: Robot Long-Term Memory Index — Graph Foundry Lab (Prof. Nick Koudas), University of Toronto, Toronto, Canada (Apr 2026 - Present). Building a searchable long-term memory for robots: objects extracted from RGB-D video, indexed by identity, 3D location and time — answering "where did I last see the keys?" style queries at 44% top-1 accuracy vs. 13% for an HNSW baseline, with only 2% added latency on small queries.
- Summer Research Intern — VineGraph — SAIL Lab, Shanghai Jiao Tong University, Shanghai, China (May 2025 - Aug 2025). Worked on VineGraph, a vector-search index that runs from SSD instead of RAM, so billion-vector datasets fit on ordinary machines.
- Software Development Intern — Exclusive Dry Cleaners, Toronto, Canada (Apr 2025 - May 2025). Maintained a Laravel dry-cleaning platform: fixed email delivery and order-status bugs affecting guest users.
Projects and Research
- Certified Memory Consolidation for LLM Agents (Mar - Jul 2026). AI agents repeatedly compress old memory, quietly destroying facts; showed the standard statistical safety check cannot predict when, and built a distribution-free alternative valid in all 108 test settings (baseline: one third).
- Football Match Prediction via Mathematical Modelling (2023). Combined bookmaker odds and team offensive/defensive metrics to build a predictive model for match outcomes.
