From Idea to Deployment: Building My Portfolio With Gemini

This is a submission for the New Year, New You Portfolio Challenge Presented by Google AI

About Me

My name is Konark Sharma, and I’m a tech enthusiast at heart. I love exploring new technology whether it’s reading about it, watching product launches, or following the latest hardware innovations. I enjoy staying up to date with what’s happening in the tech world and discussing new ideas.

I’m a curious learner who enjoys building and experimenting. Professionally, I’m a React developer, and recently I’ve started learning Go, which I’ve found to be a fun and refreshing language.

This portfolio reflects my curiosity, learning mindset, and growing interest in AI-powered development.

Portfolio

How I Built It

I built this portfolio using Google AI Studio with an iterative, prompt-driven approach.

Tech Stack

• Frontend: React, TypeScript
• Backend & Deployment: Google APIs, Google Cloud Run
• AI Model: Gemini (2.5 Flash → Gemini 3)

I started with Gemini 2.5 Flash because of its lower cost, but I wasn’t fully satisfied with the results. The responses required more manual corrections, and errors tended to repeat. Switching to Gemini 3 significantly improved the experience its faster reasoning, stronger problem-solving, and better error handling made development much smoother.

The process felt almost like translating ideas directly into working code. Gemini handled links, layout suggestions, and many implementation details with minimal friction.

That said, working with AI wasn’t always smooth. At times, the model behaved unpredictably failing to remove files, hallucinating features, or re-running older tasks alongside new ones. I often had to roll back to previous versions and manually refactor the code to regain control.

Despite these challenges, I was able to build a portfolio that closely matches my original vision being minimal, fast, and focused on small interactions. The hands-on debugging and refactoring taught me a lot about working effectively with AI tools.

What I'm Most Proud Of

• The loading animation, which sets the tone for the site
• The skills section, where I was able to execute a unique idea, even if it’s still evolving
• A minimal theme with consistent colors and typography
• Smooth, fast animations that improve the overall experience
• Most importantly, turning an idea into a real, working product
• Gaining practical experience with Gemini 2.5 and Gemini 3 something I’d love to write about in more depth later Working with AI can be challenging, but it’s also incredibly rewarding.

Thanks for checking out my submission.

I’d love to hear your feedback so I can make the next version even better.