This is a submission for the [New Year, New You Portfolio Challenge Presented by Google AI](https://dev.to/challenges/new-year-new-you-google-ai-2025-12-31
Note: I’m not a frontend developer, but I’m confident this portfolio won’t disappoint—and it reflects how AI will redefine the way engineers build, operate, and collaborate in the future 🚀
(https://aravind-927658424578.us-central1.run.app)
About Me
I’m a Cloud & SRE Engineer with deep expertise across Azure, AWS, and GCP, specializing in automation-first infrastructure, Kubernetes platforms, and AI-driven operations.
I focus on building self-healing, governance-driven, and scalable cloud systems using Terraform, CI/CD pipelines, and agentic AI frameworks. This portfolio reflects how I think as an engineer—starting from fundamentals, scaling through automation, and delivering measurable real-world impact.
Portfolio
Developing a high-impact developer portfolio requires more than just clean HTML; it requires an AgenticAI Experience. In this project, I built Quantum Folio—a developer portfolio—using an ADK (Agentic Development Kit) approach, A2A (Agent-to-Agent) communication, and Dockerized deployment on Google Cloud Run, orchestrated by the Antigravity agentic, Stitch and powered by GeminiCLI.
Guide for Realm
Upon entry, an AI agent named Gravion greets you and helps you explore the portfolio. You can either select from pre-built questions or craft your own prompts—for example: “Tell me about your creator” or “Tell me about your creator’s experience” 💬✨.
The main page highlights skills and experience, while the About section is accessible at the bottom of the interface.
How I Built It
🏗️ ADK Architecture (Agentic Development Kit)
Instead of a static website, this portfolio is built as a modular ADK, allowing new capabilities to be added through independent agents.
🔄 A2A (Agent-to-Agent) Flow
- Router Agent – Interprets user intent
- Skills Agent – Technical stack and expertise
- Experience Agent – Professional journey
- Feedback Agent – User feedback and interaction
- Quantum Agent – Physics-inspired explanations
This A2A design ensures the right “expert” agent responds every time, improving both UX and scalability ⚙️🧠.
🛠️ The Engine: Antigravity & Gemini CLI
This project wasn’t just coded—it was co-piloted.
- Gemini CLI enabled creative reasoning, long-context understanding, agent logic, and domain knowledge.
- Antigravity handled infrastructure decisions, Docker configuration, and Cloud Run troubleshooting.
Together, they enabled rapid iteration while maintaining architectural clarity 🚀.
Security
Agentic Security: A2A Rate Limiting
To protect the ADK from abuse, I implemented agent-level rate limiting using a sliding time window:
checkRateLimit(channel) {
const history = this.rateLimits[channel] || [];
const recent = history.filter(ts => Date.now() - ts < 60000);
if (recent.length >= 3) return false; // Security Alert triggered
recent.push(Date.now());
this.rateLimits[channel] = recent;
return true;
}
What I'm Most Proud Of
⚛️ Turning a portfolio into an agentic system, not a static site
🧠 Applying A2A patterns to a real UX problem
☁️ Seamless Cloud Run + Docker deployment
🎨 A unique Quantum Zero-G visual metaphor that reflects modern distributed systems
🚀 Demonstrating how Google AI + agentic tooling can elevate developer experiences
A glimpse into QuantuME Folio
Thanks to Google for providing this opportunity and inspiring me to step outside my comfort zone and learn more about frontend development.