ITEN

Emiliano Pisu

Emiliano Pisu

Senior Frontend Engineer ✨ AI Product Engineer 🎨 Design Engineer ♿️ Accessibility WCAG Advocate 🎤 Speaker 👨🏻‍🏫 Mentor

Bio

Hi 👋 Folks! I’m Emiliano Pisu, your friendly neighborhood Design Engineer. Throughout my career, driven by my coding addiction, I’ve always stayed up to date with software development and related trends, focusing on UI, UX, Accessibility, Responsiveness, Compatibility, Scalability, and Maintainability. My main goal is to build accessible and pixel-perfect user interfaces that enhance the user experience. My core approach is to develop with zero dependencies, fostering creativity, learning, experimentation, and deep knowledge. This leads to solutions that not only follow trends but are also carefully tailored to the needs of clients and stakeholders. I am deeply committed to knowledge sharing and mentorship: through my experience as Sensei @ Dev Dojo IT, I actively support those who want to grow in the frontend world and in designing accessible and scalable interfaces. In conclusion, I’m the right person if you need a bridge between Design and Development. I live to simplify my colleagues' DevX, which is why I focus on advocating for the core technologies of the web, strictly vanilla.

Sessions

SAT 28 NOV

11:35

12:20

Monterosa

LaunchpadIntermediateeng

Deterministic MCP Server

Emiliano Pisu - Senior Frontend Engineer ✨ AI Product Engineer 🎨 Design Engineer ♿️ Accessibility WCAG Advocate 🎤 Speaker 👨🏻‍🏫 Mentor

What happens when a local MCP server is designed to enforce engineering standards rather than simply exposing tools? After six months of experiments, I developed and rigorously tested `pix-galaxy-mcp`, an Agent-Skills-focused MCP server featuring 235 tools, 146 prompts and resources, 70 indexed skills, and a deterministic control engine covering frontend, backend, accessibility, design systems, testing, architectural decision records, and cross-session handoffs. It was used in four real-world projects—implemented using the same prompt but across three different models—resulting in 12 measurement sessions comparing Claude Sonnet, GPT, and DeepSeek. The results challenge several common assumptions: the MCP server accounted for ~5% of the total project effort, and prompt caching significantly altered costs. This presentation offers a deep, honest technical analysis backed by billed token usage, generated code, MCP compliance scores, architectural outcomes, and failed experiments. It includes a practical demonstration of safeguards, such as automatic fail-safe shutdowns and deterministic validation. The MCP server does not replace model quality, but it can transform vague coding conventions into measurable engineering safeguards.