About Me

🎯 What I Do

I specialize in turning traditional test automation discipline into agentic AI reliability engineering. My work, which began with published research into spatial data clustering, now focuses on Python tooling, local-first AI systems, LLM test harnesses, run replay, and failure diagnostics for workflows where traditional automation alone is no longer enough.

I build tools that do not just run scripts, but make complex work inspectable: self-healing automation, deterministic LLM tests, agent flight recorders, and postmortems that explain what happened before a team trusts the next run.

💡 My Philosophy

I do not just automate workflows; I engineer the evidence layer that makes automation and AI-agent behavior safer to trust.

"Reliable autonomy starts with observable execution."

📚 Research Foundations

My approach to engineering is rooted in a background in data mining and algorithmic research. Before moving into automation, I co-authored "Clustering Concepts and Techniques - For Big Spatial Data", focusing on the Triangle-Density Based Clustering Technique (TDCT).

Today, I apply those same principles of pattern recognition and data clustering to solve complex problems in test stability and self-healing automation frameworks. You can find a record of my academic research and citations on Google Scholar.

🛠️ Core Expertise

☕ Fun Fact

I like the kind of engineering where a vague failure becomes a reproducible story. When I am not building automation and AI reliability tools, I am usually exploring local AI systems, writing technical guides, or polishing open-source utilities.