title: Boniface.dev — Mission Control AI Portfolio Built with Gemini & Cloud Run
published: true
tags: googleai, gemini, ai, cloudrun, portfolio
label:dev-tutorial=devnewyear2026
dev-tutorial: devnewyear2026
This is a submission for the New Year, New You Portfolio Challenge Presented by Google AI
About Me
Hi! I’m Boniface (Bon) — an AI Architect & Engineer focused on building production-grade AI systems that operate reliably at scale. My passion lies in designing and deploying advanced GenAI systems, agent orchestration frameworks, and retrieval-augmented workflows — all with strong attention to safety, clarity, and real-world usability.
Rather than listing technologies, I focus on solving real engineering problems — optimizing for performance, reliability, and real-world constraints. My portfolio reflects that mindset.
Portfolio
Here’s my live portfolio — deployed on Google Cloud Run as required for this challenge:
🤖 Nexus AI — Interactive Portfolio Guide
This portfolio includes Nexus AI, an AI-powered interaction layer built using Google Gemini.
Nexus AI allows reviewers to:
- Ask questions about any project
- Get architecture explanations in plain English or deep technical detail
- Navigate the portfolio intelligently instead of manually scanning
- Understand trade-offs, decisions, and system design choices
Example prompts reviewers can try:
- “Explain this portfolio like I’m a Google AI judge”
- “Walk me through your RAG architecture”
- “What problem does your agent framework solve?”
This feature is designed to help reviewers understand not just what I built — but why.
💡 Tip for reviewers: Try asking Nexus AI “Explain this portfolio like I’m a Google AI judge.”
How I Built It
🧰 Tech Stack
- Frontend: React / Next.js
- Backend: Node / FastAPI-style services
- AI Integration: Google Gemini via Cloud Run service account + IAM
- Deployment: Google Cloud Run (serverless)
- Containerization: Docker
🛠 Design & Development Approach
- Purpose-first design: The UI uses a “Mission Control” metaphor to communicate my approach as an engineer — systematic, intentional, and operationally grounded.
- Secure AI usage: I access Gemini server-side only using IAM & application default credentials (no exposed API keys).
- Scalable deployment: Cloud Run provides reliable, autoscaled hosting with HTTPS built in and minimal ops overhead.
- Responsive layout: The interface adapts to different screen sizes and focuses on readability and discovery.
🧠 Google AI Tools Used
- Gemini models for backend contextual insights and content exploration
- Google Cloud Run for secure hosting
- Service Account IAM for authenticated AI access
- Google Antigravity : IDE
What I'm Most Proud Of
💡 Innovation & Technical Implementation
- Integrated Gemini in a secure, backend-only fashion
- Designed portfolio UI that reflects an engineering mindset
- Deployed on Google Cloud Run with autoscaling and HTTPS
🚀 User Experience
- Fast, accessible navigation
- Clear project storytelling and context
- Mission Control theme that ties design and function
🎯 Demonstrated Skills & Depth
This portfolio isn’t just a showcase — it’s a living system demonstrating real architectural decisions and scalable cloud deployments, which directly reflects how I build production AI systems.
Thanks for reviewing my submission, and thank you to the Google AI team for hosting this challenge! 🚀
Top comments (0)