
Ivan Siianytsia
Ivan leads the Connect & Integrations team at Metorial, focusing on creating seamless connections between our platform and various enterprise tools. He has a background in software engineering and focuses on reliability and adaptability in complex systems.
Blog Posts
Security
The prompt injection hiding in your tickets and emails
Agents read text written by strangers as part of normal work. You cannot reliably filter out every instruction, so limit what an agent can do after it reads one.
Engineering
Don't let your agent improve the code
Part 2 of Agents on the team. This time I forced coding agents to follow each repo directly, built a skill to investigate production bugs, and shortened the wait for CI.
Security
The first hour after an AI agent makes a damaging change
A practical response sequence for containing an agent, preserving evidence, finding the full scope, and recovering without losing the trail.
