New Year, New You: Architecting a High-Performance AI Portfolio using Python and Gemini

Source: Dev.to

Overview

This submission for the New Year, New You Portfolio Challenge showcases a “System Architect’s Console”—a live, gamified environment that demonstrates how a senior‑level perspective can leverage Google AI to solve enterprise‑scale problems such as compliance and developer productivity.

Tech Stack

Backend

• Python 3.12 (Standard Library only) hosted on Google Cloud Run.

Frontend

• Vanilla JavaScript (ES6 Modules) and pure CSS with a “Midnight Architecture” theme.

AI Integration

• Google Antigravity used as the primary development tool.
• Gemini AI: Integrated the Gemini 3 Flash Preview model via raw HTTP calls (no SDKs) to power a “Technical Consultant” persona that answers recruitment queries using project‑specific data.

Gamification

• Built a custom “System Audit” engine using the IntersectionObserver API to track user engagement and unlock technical “milestone logs” as visitors explore the career path.

Performance & Security

• Achieved a 95+ PageSpeed score while maintaining a fully interactive, AI‑powered experience.
• The License Compliance Agent feature bridges BigQuery and Generative AI, reducing manual audit effort by 70 %.
• Implemented with zero third‑party libraries and zero critical vulnerabilities, adhering to security‑first principles aligned with ISO 27001 and SOC 2 audit standards.