SangYoon Yu
Entrepreneur • AI Security Researcher • Align Human & Tech
___________
| ___ |
| |o o| |
| |>_<| |
| |===| |
|___________|
/| [o] |\
/ | || | \
/ | || | \
|_________|
| | | |
_| | | |_
|____| |____|
___________________
| _____________ |
| | | |
| | HELLO | |
| | WORLD! | |
| |_____________| |
|___________________|
|_____[_]_[_]_[_]___|
|_____|
| |
/| |\
// About Me
About
Hello World. I'm the CEO and Co-founder of AIM Intelligence, where we build safety and security systems for the AI era. I earned my M.S. in Electrical and Computer Engineering at Seoul National University under Professor Soo-Mook Moon, focusing on Responsible AI, Federated Learning, and AI Fairness—the foundation of my ongoing work in AI safety.
I focus on making AI systems robust and trustworthy in real-world scenarios—even for high-risk tasks:
- How can we design AI systems that can be reliably delegated high-stakes responsibilities?
- How can we ensure models remain robust against adversarial attacks from malicious actors?
Interested in: AI Safety • AI Security • Responsible AI • Adversarial Attacks • AI Guardrails
// Projects
master
│
●
╱│╲
╱ │ ╲
╱ │ ╲
● │ ●
feature │ hotfix
│
●
merge
Work
AIM Guard
In ProductionReal-time AI safety layer. Custom policies, multimodal support, enterprise-ready.
AIM Red
In ProductionAutomated red-teaming platform. Continuous adversarial testing and safety audits.
Korean PII Dataset
Open SourceLarge-scale PII detection corpus. 100K+ annotated samples for Korean language.
National AI Security Challenge
Dec 1-2, 2024Korea's flagship AI security event hosted with the government.
// Publications
Research
_______________
/ \
|_________________|
|| ||
|| DATABASE ||
|| ||
|_________________|
|| ||
|| STORAGE ||
|| ||
|_________________|
\_______________/
// Contact
[NODE]
/ | \
/ | \
/ | \
[A]----[B]----[C]
\ | /
\ | /
\ | /
[HUB]
Contact
For research collaborations and speaking:
team@aim-intelligence.comBased in Seoul, Korea