How I Deployed My AI/ML Portfolio in Under 5 Minutes Using Kuberns AI

Ananya — Tue, 03 Mar 2026 11:42:24 +0000

A developer's honest experience with one-click AI deployment

A bit about me

I'm Ananya — an Applied AI Engineer focused on Computer Vision, ML Pipelines, and Intelligent Automation. I spend most of my time building end-to-end AI systems: training CNN models, building CV pipelines with OpenCV and TensorFlow, and packaging everything with Docker.

What I'm not great at? Deployment infrastructure. DNS configs, server setup, build pipelines — it's a whole other job. So for a long time, my projects lived on GitHub and nowhere else.

That changed when I entered the Kuberns AI Portfolio Hackathon 2026.

The Problem: Great Projects, No Live URL

I had built real AI systems:

EcoStream AI — environmental intelligence system analyzing real-time AQI data
Automated OMR Evaluation — computer vision pipeline for automated exam scoring
Face Mask Detection — real-time CNN-based safety monitoring system
Foodoscope — deep learning food image classifier
Financial Chatbot — NLP-powered finance assistant

All of them sat on GitHub. None had a live URL. And in 2026, if your project isn't live for most recruiters, it doesn't exist.

Entering the Hackathon

The Kuberns AI Portfolio Hackathon 2026 had one clear goal: deploy a live portfolio. Not build something from scratch. Not write a thesis. Just ship it, make it public, show proof.

The prizes were real:

Top 5 share a $10,000 prize pool
Top 30 get featured permanently on Kuberns' social channels
Top 100 get Amazon Pay gift cards
Internship/Full-time opportunity up to 10 LPA

But honestly, the bigger reward was finally having a live portfolio URL I could put on my resume.

Building the Portfolio

I built a single-page HTML portfolio with:

Dark theme with animated particle network (canvas-based, mouse-reactive)
Custom cursor with magnetic buttons
Scroll-reveal animations on every section
Architecture diagrams for each project showing the system pipeline
Skills grid organized by category
Contact section with GitHub, LinkedIn, and email links

The design philosophy: make it look like a product page, not a student assignment.

Every project card follows the same formula:

Project Name
Short problem → solution description
Tech stack tags
Architecture pipeline: Input → Processing → Model → Output
GitHub link + Demo link

That architecture pipeline was the single biggest upgrade. It signals systems thinking instantly to any engineer reading the portfolio.

Deploying on Kuberns

Here's where it got interesting.

I expected deployment to be the hard part. It wasn't.

Steps I took:

Went to kuberns.com
Signed up (free)
Uploaded my index.html file
Clicked deploy

That's it.

No Dockerfile to write. No environment variables to configure. No server to provision. Kuberns AI handled the build, deployment, and hosting automatically.

My portfolio went live in under 5 minutes.

Live portfolio: https://ananya2306-portfolio-main-9ded0fe.kuberns.cloud/

What Makes Kuberns Different

As someone who has manually deployed on Vercel, Render, and set up Docker containers, the difference with Kuberns is the AI layer.

It doesn't just host files. It understands what you're deploying and configures the environment accordingly. For a static portfolio, that meant instant deployment. For more complex apps, it handles the build pipeline intelligently.

The one-click promise is real. I was skeptical. I'm not anymore.

Lessons from Building a Strong Portfolio

After multiple rounds of feedback and iteration, here's what actually matters:

1. Engineer your descriptions, don't just describe

❌ "Built a mask detection model"
✅ "Designed a computer vision pipeline for real-time mask compliance detection using CNN-based classification"

Same project. Completely different perception.

2. Architecture diagrams are gold

Video Feed → Face Detection → CNN Classifier → Alert System

Four words and three arrows. Engineers instantly understand the system. Add this under every project.

3. Deployment proof > everything

A GitHub link says "I wrote code."
A live URL says "I ship products."

That difference matters enormously to recruiters.

4. Organize skills by category

Don't dump 20 technologies in a list. Group them:

Programming
ML & AI
CV Libraries
Deployment
Engineering Tools

Recruiters scan in 8 seconds. Make it easy.

Final Thoughts

If you're a developer with projects sitting on GitHub and no live portfolio — just deploy it. The Kuberns hackathon is the perfect forcing function.

The hardest part isn't the code. It isn't even the design. It's just shipping.

Kuberns made the shipping part trivially easy. The rest was up to me.

Links:

🌐 Live Portfolio: https://ananya2306-portfolio-main-9ded0fe.kuberns.cloud/
💻 GitHub: https://github.com/Ananya2306
🏆 Hackathon: kuberns.com
💬 Discord: https://discord.gg/KRCM9QneTJ

Built and deployed as part of the Kuberns AI Portfolio Hackathon 2026

#kubernschallenge #kuberns #aichallenge #MachineLearning #ComputerVision #Deployment #Portfolio

𝐃𝐞𝐞𝐩 𝐋𝐞𝐚𝐫𝐧𝐢𝐧𝐠: 𝐀 𝐓𝐢𝐦𝐞𝐥𝐢𝐧𝐞 𝐨𝐟 𝐊𝐞𝐲 𝐌𝐢𝐥𝐞𝐬𝐭𝐨𝐧𝐞𝐬

Ananya — Thu, 15 Jan 2026 16:26:27 +0000

Deep learning has evolved over several decades through continuous advances in neural network models, learning algorithms, and computational power.
The timeline below highlights the key milestones and contributors that shaped modern deep learning.

1943 -- 𝐍𝐞𝐮𝐫𝐚𝐥 𝐌𝐨𝐝𝐞𝐥 (𝐌𝐜𝐂𝐮𝐥𝐥𝐨𝐜𝐡 & 𝐏𝐢𝐭𝐭𝐬)
• First mathematical model of a biological neuron
• Foundation of artificial neural networks

1957 -- 𝐏𝐞𝐫𝐜𝐞𝐩𝐭𝐫𝐨𝐧 (𝐅𝐫𝐚𝐧𝐤 𝐑𝐨𝐬𝐞𝐧𝐛𝐥𝐚𝐭𝐭)
• First learning algorithm for neural networks
• Enabled binary classification using weighted inputs

1982 -- 𝐇𝐨𝐩𝐟𝐢𝐞𝐥𝐝 𝐍𝐞𝐭𝐰𝐨𝐫𝐤 (𝐉𝐨𝐡𝐧 𝐇𝐨𝐩𝐟𝐢𝐞𝐥𝐝)
• Introduced recurrent neural networks
• Enabled associative memory

1985 -- 𝐁𝐨𝐥𝐭𝐳𝐦𝐚𝐧𝐧 𝐌𝐚𝐜𝐡𝐢𝐧𝐞 (𝐇𝐢𝐧𝐭𝐨𝐧 & 𝐒𝐞𝐣𝐧𝐨𝐰𝐬𝐤𝐢)
• Introduced stochastic learning
• Basis for deep representations

1986 -- 𝐁𝐚𝐜𝐤𝐩𝐫𝐨𝐩𝐚𝐠𝐚𝐭𝐢𝐨𝐧 (𝐑𝐮𝐦𝐞𝐥𝐡𝐚𝐫𝐭, 𝐇𝐢𝐧𝐭𝐨𝐧, 𝐖𝐢𝐥𝐥𝐢𝐚𝐦𝐬)
• Enabled training of multilayer networks
• Core optimization algorithm

𝐋𝐚𝐭𝐞 1980𝐬 -1990𝐬 -- 𝐀𝐈 𝐖𝐢𝐧𝐭𝐞𝐫
• Limited computation and reduced funding
• Shift toward simpler ML models

1990 -- 𝐒𝐕𝐌 (𝐕𝐥𝐚𝐝𝐢𝐦𝐢𝐫 𝐕𝐚𝐩𝐧𝐢𝐤)
• Margin-based classification
• Effective for high-dimensional data

1997 -- 𝐋𝐒𝐓𝐌 (𝐇𝐨𝐜𝐡𝐫𝐞𝐢𝐭𝐞𝐫 & 𝐒𝐜𝐡𝐦𝐢𝐝𝐡𝐮𝐛𝐞𝐫)
• Solved vanishing gradient problem
• Enabled sequence learning

𝐋𝐚𝐭𝐞 1990𝐬 - 2000𝐬 -- 𝐆𝐏𝐔 𝐂𝐨𝐦𝐩𝐮𝐭𝐢𝐧𝐠
• Accelerated neural network training
• Enabled large-scale deep learning

2012 -- 𝐃𝐞𝐞𝐩 𝐋𝐞𝐚𝐫𝐧𝐢𝐧𝐠 𝐑𝐞𝐯𝐨𝐥𝐮𝐭𝐢𝐨𝐧 (𝐀𝐥𝐞𝐱𝐍𝐞𝐭)
• CNN breakthrough on ImageNet
• Widespread DL adoption

2014 -- 𝐆𝐀𝐍𝐬 (𝐈𝐚𝐧 𝐆𝐨𝐨𝐝𝐟𝐞𝐥𝐥𝐨𝐰)
• Introduced adversarial learning
• Enabled data and image generation

2016 -- 𝐃𝐞𝐞𝐩 𝐑𝐋 (𝐀𝐥𝐩𝐡𝐚𝐆𝐨)
• DL + reinforcement learning
• Superhuman decision-making

2017 -- 𝐓𝐫𝐚𝐧𝐬𝐟𝐨𝐫𝐦𝐞𝐫𝐬 (𝐀𝐬𝐡𝐢𝐬𝐡 𝐕𝐚𝐬𝐰𝐚𝐧𝐢 𝐞𝐭 𝐚𝐥.)
• Introduced self-attention
• Foundation of modern NLP

2020 -- 𝐒𝐞𝐥𝐟-𝐒𝐮𝐩𝐞𝐫𝐯𝐢𝐬𝐞𝐝 𝐋𝐞𝐚𝐫𝐧𝐢𝐧𝐠 (𝐘𝐚𝐧𝐧 𝐋𝐞𝐂𝐮𝐧)
• Reduced need for labeled data
• Improved representation learning

2022 - 𝐏𝐫𝐞𝐬𝐞𝐧𝐭 -- 𝐋𝐚𝐫𝐠𝐞 𝐋𝐚𝐧𝐠𝐮𝐚𝐠𝐞 𝐌𝐨𝐝𝐞𝐥𝐬
• Scaled transformers for multimodal AI
• Real-world deployment at scale

DEV Community: Ananya