DEV Community: Gulrez Qayyum

GiveBack — Turn Generosity Into Action

Gulrez Qayyum — Sun, 06 Sep 2026 16:01:55 +0000

This is a submission for Weekend Challenge: Generosity Edition

What I Built

GiveBack is an AI-powered platform designed to make generosity easier and more actionable.

The idea is simple: people often want to help others, but don't always know how they can contribute, where their help is needed, or what meaningful action they can take.

GiveBack uses AI to turn a user's intention to help into practical suggestions and actions.

Instead of simply asking:

"How can I donate?"

GiveBack focuses on:

"What can I do today that actually helps someone?"

The goal is to encourage everyday generosity — whether that's donating resources, volunteering time, helping someone in the community, or contributing in another meaningful way.

I wanted to build something that demonstrates that AI doesn't always need to be about productivity or automation. It can also be used to encourage people to help each other.

Demo

Live Application: https://give-back-five.vercel.app/

Code

GitHub: https://github.com/GulrezQayyum/give-back

The project contains the frontend and backend components of GiveBack.

How I Built It

I built GiveBack with a simple full-stack architecture:

Frontend

The frontend was built with React + Vite.

It provides the user interface where people can interact with GiveBack and describe how they would like to contribute.

Backend

The backend was built with Python and FastAPI.

The backend handles application logic and acts as the bridge between the frontend and the AI functionality.

Google AI

The core AI functionality uses Google AI / Gemini.

Gemini is used to understand the user's intent and generate meaningful, context-aware suggestions for ways they can give back.

Rather than returning generic suggestions, the goal is to make the interaction feel like a conversation:

User intention → AI understands the situation → GiveBack suggests meaningful actions

Why AI?

Generosity is highly contextual.

Someone might have:

time but no money
money but limited time
technical skills they want to contribute
resources they don't need anymore
a desire to help their local community

AI provides a natural way to understand these different situations and suggest possibilities based on what a person is actually able and willing to give.

Prize Categories

Best Use of Google AI

GiveBack is submitted for Best Use of Google AI.

I integrated Google Gemini into the application to understand user intentions and generate personalized suggestions for meaningful acts of generosity.

The AI isn't being used simply as a chatbot. It is part of the core experience: helping users move from "I want to help" to "Here's something meaningful I can actually do."

Why I Built This

The idea behind GiveBack came from a simple observation:

We often think generosity requires money.

But generosity can also be time, knowledge, skills, attention, resources, or simply helping someone who needs it.

I wanted to experiment with how AI could help people discover those opportunities.

GiveBack is my attempt to build technology around a more human goal:

Use AI not only to help us do more, but to help us give more.

RAGBench: Stop Guessing, Start Measuring Your RAG Pipeline

Gulrez Qayyum — Wed, 02 Sep 2026 16:43:14 +0000

Hey builders

RAGBench just hit Hugging Face.

It's an open evaluation framework for RAG systems, built for developers, with the goal of making RAG evaluation more practical and reproducible.

Instead of guessing whether your chunking strategy actually works, test it. Compare it. Measure it.

I built RAGBench to make it easier to evaluate different RAG approaches using the same documents, queries, and evaluation metrics.

So far, I've focused on comparing semantic, document-level, and parent-child chunking and measuring how they affect retrieval and answer quality.

The goal is simple:

Stop guessing. Start measuring.

RAGBench is:
• Open source
• Open data
• Reproducible
• Built for community feedback

Try the interactive benchmark:
https://huggingface.co/spaces/Gul55555/ragbench

Contributions and feedback are welcome.

I'm especially interested in:
• What RAG evaluation metrics do you actually use?
• What chunking strategies have worked best for you?
• What problems have you encountered when evaluating RAG?
• What should I add to RAGBench next?

What's broken in your RAG pipeline?

I'd love to hear what you're working on.

🍝 Pasta Shape Guide — An Interactive Guide to the Pasta We Love

Gulrez Qayyum — Sat, 15 Aug 2026 16:47:55 +0000

This is a submission for Frontend Challenge - Comfort Food Edition, Perfect Landing

What I Built

For the Frontend Challenge: Comfort Food Edition, I decided to build around one of my favorite comfort foods: pasta. 🍝

Instead of creating a simple recipe landing page, I wanted to make something people could actually explore.

So I built Pasta Shape Guide, an interactive guide to different pasta shapes, their origins, cooking times, recommended sauces, recipes, and cooking tips.

You can:

🔍 Search pasta by name or sauce
🍝 Browse different pasta shapes
🌍 Learn about their origins
⏱️ Check recommended cooking times
🍅 Discover sauce pairings
📖 Read detailed recipes and cooking tips

Built With

React
Vite
Tailwind CSS
React Router
JavaScript
Vercel
Pexels for the pasta imagery

Demo

pasta-guide-with-react.vercel.app

Journey

This project was especially interesting for me because it was my first React project.

I'm more familiar with Flutter development, so moving into React gave me a chance to learn a different approach to building interfaces.

During the project, I learned about component-based architecture, React state, client-side routing, dynamic URLs, filtering data, responsive design, and deploying a production React application.

One thing I particularly enjoyed was making the application data-driven.

The pasta information is separated from the UI, so adding another pasta doesn't require creating an entirely new component. The same components can render the new information automatically.

What I'm Proud Of

I'm most proud that I turned what could have been a static landing page into an actual little interactive experience.

The user isn't just looking at a pasta image.

They can:

Discover → Search → Explore → Learn → Find a Sauce → Try a Recipe

That was the experience I wanted to create.

I'd love to know: what pasta shape would you add to the guide? 🍝

Building an AI-Powered Network Traffic Analyzer with FastAPI and Scapy (Looking for Contributors)

Gulrez Qayyum — Fri, 17 Jul 2026 12:11:12 +0000

Over the past few weeks I've been building NetGuard, an open-source AI-powered Network Traffic Analyzer and Active Defense Platform.

The goal is to combine traditional intrusion detection with machine learning into a single platform capable of:

Capturing live packets using Scapy
Detecting suspicious traffic using IDS rules
Classifying traffic with a Random Forest model
Automatically blocking malicious IPs
Providing remediation recommendations
Generating incident reports

Most of the system is already working, including packet capture, IDS detection, remediation, dashboard integration, and IP blocking.

The remaining work is mainly around integration.

The most interesting bug is that the FastAPI orchestrator successfully receives HTTP 200 OK from the classifier service, yet ai_result is always returned as null.

I've investigated payload formatting, response parsing, and model loading, but I'm convinced another pair of eyes could spot what I'm missing.

I've documented the architecture, setup instructions, debugging notes, and open issues in the repository.

If you're interested in Python, FastAPI, networking, cybersecurity, or machine learning, I'd love your feedback or contributions.

Repository:
https://github.com/GulrezQayyum/netguard

Open-source is one of the best ways to learn, so even documentation improvements or code reviews are welcome.

Apex F1 – The Ultimate Formula 1 Racing Simulator

Gulrez Qayyum — Mon, 13 Jul 2026 12:51:08 +0000

This is a submission for Weekend Challenge: Passion Edition

Demo

https://apex-f1-159.web.app

Code

https://github.com/GulrezQayyum/apex-f1

Inspiration

Formula 1 has always fascinated me because it's not just about speed—it's a sport driven by strategy, teamwork, weather, tyre management, and split-second decisions. I wanted to capture those elements in a mobile app rather than building a traditional arcade racing game. Apex F1 is my attempt to recreate the excitement of managing a race weekend while combining my passion for Formula 1 with Flutter development.

How I Built It

I built Apex F1 entirely with Flutter, focusing on creating a realistic Formula 1 racing and strategy experience for mobile devices. As a huge Formula 1 fan, I wanted to go beyond a simple racing game and recreate the tactical decisions that make the sport exciting.

The project started with designing the app's architecture and user interface in Flutter. I then developed the race simulation engine, which powers every lap of the race. The simulator considers factors such as tyre degradation, fuel strategy, weather conditions, pit stops, safety cars, and AI driver behavior to produce dynamic race outcomes.

To make each race feel unique, I implemented:

GPS-inspired circuit layouts based on real Formula 1 tracks.
Twenty AI-controlled drivers with different performance characteristics.
Dynamic weather that can change race strategy.
Tyre wear and pit stop strategy affecting lap times.
Championship standings that carry progress across an entire season.
Multiple seasons (2023–2025) with updated teams and drivers.

The UI was designed to be clean and immersive, allowing users to follow race events, monitor strategy, and track championship progress without feeling overwhelmed.

The biggest challenge was balancing realism with performance. Building a simulation that remained responsive on mobile devices required careful optimization and repeated testing. I refined the race logic over many iterations until the races felt both competitive and unpredictable.

Overall, Apex F1 combines my interests in Flutter development, simulation programming, and Formula 1 into a single project that reflects both my technical skills and my passion for the sport.

Prize Categories

Best Use of Google AI

Apex F1 integrates Google Gemini AI to enhance the racing experience beyond traditional simulation. Google AI powers intelligent race insights and assists users with strategy recommendations, making the app more interactive and engaging. By combining AI-driven assistance with a realistic Formula 1 race simulator built in Flutter, Apex F1 delivers a smarter and more immersive experience for F1 fans.

Built by: Gulrez Qayyum — an indie developer passionate about Flutter, AI, and Formula 1.

Built a Network Traffic Classifier with Random Forest (96.8% Accuracy)

Gulrez Qayyum — Sun, 17 May 2026 18:47:18 +0000

I recently completed a cybersecurity + machine learning project where I trained a Random Forest model to classify network traffic into multiple attack categories using the NSL-KDD dataset.

The classifier can detect:

DoS attacks
Probe/reconnaissance traffic
R2L brute-force attempts
U2R privilege escalation
Normal traffic
Stack
Python
Scikit-learn
FastAPI
Pandas / NumPy
Results
96.8% Accuracy
<1ms inference time
Production-ready model packaging

I also wrote a detailed Medium article covering:

Dataset preprocessing
Feature selection
Model training
API integration
Real-world deployment considerations
Challenges working with synthetic vs real traffic data

Would appreciate feedback from the community, especially from people working in:

Cybersecurity
ML Engineering
Intrusion Detection
Backend Systems

Article: https://medium.com/@gulrezqayyum/building-a-production-ready-network-traffic-classifier-with-random-forest-and-nsl-kdd-b9559346e197

GitHub:
https://github.com/GulrezQayyum

LinkedIn:
https://www.linkedin.com/in/gulrez-qayyum-999345322/