Aliaksei Hapeyeu
I'm a machine learning engineer working on AI security and end-to-end ML systems, from architecture and training to inference, evaluation, and A/B testing. I write about prompt injection, agent security, and how language models work.
Writing
- Catch the Jailbreak If You Can
Jailbreak is not a break-in. It is Frank Abagnale for neural networks: why LLMs trust the costume more than the badge.
- The Illusion of Control: Why Prompts Don't Protect AI Agents
From Permission Boundary Bypass to Dyck languages: why agent security has to live in the runtime, not in the system prompt.