Systems Thinking, Creative AI & Autonomous Infrastructure
My Story
From Creative AI Prototyping to
Agentic Infrastructure Automation
While building AI-assisted workflows for creative and generative AI products, I became deeply interested in how modern AI coding tools manage context, prompts, routing logic, and multi-step execution. My background in creative AI pushed me to think beyond simple demos: I wanted to understand how ideas can move quickly from concept, to working prototype, to deployable product experience.
Rather than treating AI tools as black boxes, I studied their workflow patterns in a controlled technical environment, focusing on context management, agentic prompt design, request orchestration, and tool-based execution. These insights helped me improve how I design AI-assisted engineering workflows, especially for fast-moving creative and automation use cases.
I then applied these learnings into my own agentic automation experiments. Using a private agentic workflow, I integrated custom tools including the UpCloud API and built an autonomous infrastructure pipeline. The system can provision a fresh Linux VPS, prepare a secured web environment behind custom middleware, test generative API endpoints, and clean up resources automatically.
This project reflects how I approach engineering: combining creative AI experimentation, backend automation, secure deployment flows, and agentic orchestration to turn technical curiosity into practical, working systems.