This is a submission for the New Year, New You Portfolio Challenge Presented by Google AI
About Me
Hi everyone! I'm Ankush Singh Gandhi, a software engineer and the "Warrior Who Codes." I specialize in full-stack development, system design, and building interactive web experiences.
But I'm not just a coder—I'm a passionate technical writer and lifelong learner. Over the last few years, my articles on Dev.to have gathered over 48,000 views, and I was previously recognized as a Top Voice in Computer Science on LinkedIn for my contributions to community discussions. I absolutely love devouring system design blogs and engineering case studies.
My goal with this portfolio was to create something that reflects this duality: a place that showcases my projects but also serves as an "Interactive Hub" for the content I create and consume. It blends a retro-terminal aesthetic with modern, high-performance UI to tell my story.
The Evolution: WarriorOS (v4.0) vs. Original
⚠️ NOTE: Work in Progress
This project is currently in active development. You are seeing a live preview of the new WarriorOS iteration.
This project is a complete reimplementation of my original portfolio, warriorwhocodes.com. While the original site served me well, I wanted to push the boundaries of what a personal site could be—moving from a static showcase to a living, breathing operating system.
� CRITICAL VIEWING EXPERIENCE RECOMMENDATION
While I have embedded the site below, I strongly prefer and recommend opening the portfolio in a New Window.
The complex terminal shortcuts, hover effects, and immersive UI are best experienced directly, not inside an iframe.
1. The New Architecture (WarriorOS)
- Status: Active Development 🚧
- Focus: Terminal-first navigation, AI integration, "One-Shot" learning modules.
2. The Original Portfolio (Legacy)
- Status: Stable (v3.0)
- Focus: Standard visual showcase.
Warrior Who Codes - Ankush Singh Gandhi
Backend & Software Developer specializing in Python, Django, Flask, and cloud systems. Also building modern web apps and Flutter-based mobile apps.
warriorwhocodes.com
How I Built It
I built WarriorOS (my portfolio) using a modern tech stack centered purely on offering a seamless user experience, accelerated by Google's AI tools.
Tech Stack
- Frontend: React (Vite), Tailwind CSS, Framer Motion for animations.
- Deployment: Google Cloud Run (Dockerized Nginx container).
- AI Integration: Google Gemini API.
- Development Partner: Built using Antigravity, Google's advanced agentic coding assistant.
Building with Antigravity & Gemini
This project didn't just use AI features; it was built with AI. I used Antigravity (powered by Gemini) to pair-program the entire "WarriorOS".
- Rapid Prototyping: I used the agent to iterate on the complex "Terminal Button" UI until it matched my exact vision—down to the pixel-perfect glow effects.
- Deployment Automation: The agent handled the containerization with Docker and the Cloud Run deployment commands, saving me hours of configuration time.
Google Gemini API Integration (The "New You" Feature)
I integrated the Gemini API directly into the site to create an AI Expert & Companion that doesn't just chat, it adapts. But more than just a chatbot, this feature represents my philosophy on Modern Content Writing: The AI + Human Symbiosis.
The Mindset: Evolution of Technical Learning
In the past, technical writing was static. You wrote a blog, someone read it, and if they got stuck, they had to leave your site to Google the answer.
I believe the future of content is interactive and adaptive.
- The Human Role: My job is to curate the structure, the roadmap, and the core insights (the human experience).
- The AI Role: Gemini's job is to be the personalized expert that adapts this content specifically for you.
With the new Personalization Profile engine, the AI doesn't just "answer questions" — it knows who you are. An Aspiring Backend Engineer learning Redis needs different analogies than a Senior Frontend Dev or a Product Manager. By letting you configure your Occupation, Goal, and Skills, WarriorOS transforms static documentation into a bespoke mentorship session.
Where is Gemini Integrated?
-
Context-Aware AI Personalization:
- Open the AI Toolbox via the floating button.
- Click [CONFIGURE AI] to set your profile (e.g., Occupation: "Student", Goal: "System Design").
- The Magic: Now, every AI tool—from the Pathfinder to the Tutor—will reshape its responses to fit your background. The Tutor uses your specific skills in examples, and the Quiz adjusts its difficulty automatically.
-
The "One-Shot" Expert:
- Navigate to the Library and open any "One-Shot" roadmap.
- The AI Expert uses Gemini 2.5 Flash to provide responses that are not just accurate but streamed in real-time for a natural conversational feel. It holds the full context of the roadmap you're reading, letting you ask deeper questions without losing scope.
-
Global AI Assistant:
- Throughout the site, the AI acts as a guide to my portfolio. It can explain the architecture of my projects, suggest which blog post you should read based on your interests, or even quiz you on the skills listed in my "About Me" section.
-
Active Recall Quizzes:
- Inside the Expert window, you can switch to "Quiz Mode." Gemini generates rapid-fire questions based on the content you're viewing, helping you verify your understanding in real-time.
What I'm Most Proud Of
- The Global Terminal System:
I didn't want the terminal to be a fake GIF. It's a fully functional global navigation CLI. No matter where you are on the site, you can hit the button and type commands like:
-
system.archive_access-> Instantly navigates to my library/blog section. -
system.profile_bio-> Jumps to the About Me section. -
python-> Launches a simulated Python environment (tryimport antigravity!). It treats the entire website navigation like a file system, giving power users a faster way to explore.
-
Semantic Terminal Navigation:
This is where the magic happens. I integrated Google Gemini's capabilities directly into the CLI. You aren't limited to memorizing strict commands.
* Ask naturally: Show me what you did at Desi Diaries
* Express intent: I want to see your backend projects
* Get context: Who are you?
The terminal intelligently parses your intent. If you want to navigate, it executes the system command. If you have a question about my work experience, it references my portfolio data to give you a concise, terminal-style answer. It transforms the CLI from a tool for power users into a conversational interface for everyone.
The "Configure AI" Ecosystem:
I built a dedicated Settings Configuration UI where users can input their occupation and goals (e.g., "Student learning DSA"). This isn't just a form—it's the brain of the site. It stores your profile locally and dynamically injects it into every single AI interaction, making the "AI Expert" feel like a persistent companion rather than a stateless bot.The Personalized Mentorship Hub:
Reflecting my passion for technical writing, I built interactive roadmap pages for topics like Redis, Django ORM, and MySQL. But instead of just static guides, I've created a hub where the AI Expert acts as a personal mentor, analyzing my written content through the lens of your career goals and occupation.Seamless Cloud Deployment:
Getting the app Dockerized and running on Google Cloud Run was a huge win. The site loads instantly, is secured with HTTPS by default, and can scale up if my blog post goes viral.
I'm excited to share WarriorOS with the community. It represents my journey from writing articles to building an entire operating system for my work.
Source Code
You can explore the full source code for WarriorOS on GitHub:
🚀 Overview
WarriorOS is a complete reimplementation of the warriorwhocodes.com portfolio. It moves away from static web design towards an "Interactive Operating System" aesthetic, blending a retro-terminal CLI with modern, high-performance UI components.
Built as a submission for the New Year, New You Portfolio Challenge, this iteration focuses on human-AI symbiosis in technical education.
✨ Key Features
- 🌐 Global Terminal (Warrior_OS CLI): A fully functional command-line interface used for navigation and system interaction.
- 🤖 AI Expert & Companion: Integrated Google Gemini API that acts as a context-aware tutor for roadmaps and technical content.
- 🗺️ One-Shot Roadmaps: Interactive, deep-dive learning modules on topics like Redis, MySQL, and Django ORM.
- ⚡ High-Contrast Aesthetics: A "Paper Terminal" theme featuring sharp black cards over white grid backgrounds for maximum readability.
🛠️ Tech Stack
- Frontend: React…
Happy Coding! 🚀
Top comments (0)