AI Engineer
Building infrastructure for AI systems and real-world intelligence. Focused on products that ship.
Projects
AI Classroom
Problem: Classrooms need a clearer way to understand attendance, engagement, and student wellbeing.
Solution: Built a real-time camera-based system that helps monitor attendance, attention, and visible emotional cues during class.
Impact: Turns classroom camera feeds into insights about attendance, attention, and student engagement.
Self-Hosted AI Assistant
Problem: AI workflows are often scattered across chat apps, file tools, and disconnected services.
Solution: Built Saba as a self-hosted AI assistant for chatting with models, working with files, using approved tools, remembering useful context, and connecting chat channels.
Impact: Gives users one private, customizable place to run AI workflows on their computer or server.
AI Agent Fiewall
Problem: AI agents need a safety and policy layer before they can operate in real systems.
Solution: Shaped an SDK concept for securing agent behavior, decisions, and execution boundaries.
Impact: Positions Megent as infrastructure for trustworthy agent workflows.
AI Classroom Representative Agent
Problem: Classroom coordination still depends on manual updates and scattered communication.
Solution: Built repIt as an AI classroom representative agent to organize messages, reminders, and coordination.
Impact: Gives classroom communication a reliable agent layer.
Low-Resource NLP
Problem: Berta-English translation has limited high-quality machine learning support.
Solution: Trained a low-resource translation pipeline for Berta language preservation and access.
Impact: Supports indigenous-language digitization and future NLP tooling.
Skills
Contact
Reach out for AI products, infrastructure work, or founder-level execution.
Email: mikilezen@gmail.com
.jpg&w=96&q=75){kind=small}