Context:
A couple of weeks ago, I took part in an exciting 8-hour hackathon organized by Cerebras, spotlighting their powerful new LLAMA 4 LLM deployment. The event hosted on Lu.ma challenged developers to build innovative AI-powered apps fast.
Spoiler alert: I didn’t win. But the project that did was seriously impressive a Satellite Signal Log Analyzer that parsed satellite radio logs in real time, with interference risk scores, visual trend charts, log comparisons, and more, all powered by Llama 4 and a sleek orange/white UI. Totally deserved the win.
As for me, I’ve been a programmer analyst for over 15 years, building web apps and CRM systems across a wide range of tech stacks Python, Next.js, PHP, Java, you name it. I’ve tackled everything from full-stack dev to backend architecture. But even with that experience, hackathons are a different kind of beast. Fast pace. High pressure. No time to overthink.
My idea was to develop an AI-powered summarizing blog app for kids, a fun, educational platform that takes regular blog content and simplifies it for younger readers using LLMs. I knew I wanted something lightweight, easy to deploy, and most importantly, user-friendly for kids.
But the clock was ticking, and I wasn’t sure where to start.
That’s when I reached for my secret weapon: TasksForge.ai. It helped me break the project into manageable chunks, map out the architecture, and stay focused throughout the sprint. Combine that with a little “vibe coding” magic, and I was off and running.
🛠️ How I Did It: From “Petit Prince” to Project Plan with TasksForge
Once I locked in my idea: an AI-powered blog summarizing app for kids. I knew I needed to quickly shape it into something concrete. I even had the perfect name in mind: “Petit Prince”. Inspired by the timeless tale, I wanted the app to be equally poetic, thoughtful, and child-friendly, something that both parents and kids could use together.
I had a rough concept: parents would paste a blog link or article into the app, select their child’s age range, and let the AI assistant rewrite the content in a simplified, kid-friendly tone. The core of the app would be powered by Llama 4, through Cerebras’s inference API.
I also envisioned a gentle, intuitive UI accessible enough for a child, but practical for a parent. With only hours to build, I needed clarity. So I turned to my secret productivity engine: TasksForge.ai, using this prompt:
As a parent, I want to share a blog post or article with my child, using the app AI assistant to summarize and rewrite the blog/article to be read by a kid. As a parent, I want to have an input text box or paste a blog link, and radio options to choose the child's age range. I want to use Llama 4 LLM provided by the Cerebras inference platform.
TasksForge.ai helped me unpack the idea, generate structured user stories, and tailor the implementation to fit my preferred tech stack. After a few iterations and a back-and-forth conversation with the TasksForge.ai AI assistant tool, I was able to refine the core requirements and user stories, aligning them perfectly with the vision in my head.
I also made sure to scope the project to the tech stack I’m most comfortable with, focusing on quick deployment and simplicity. I had a clear architecture, scoped features, and a realistic build plan all within minutes. You can even explore my exported project board here:
👉 Petit Prince Hackathon Project Board
💡 Choosing a Platform: Why I Went with val.town
For deployment, I wanted something fast, free, and frictionless. That’s when I chose val.town, a lightweight, serverless environment for deploying JavaScript/TypeScript apps. I was curious to try it, and it turned out to be a pleasant surprise. It gave me quick, no-hassle deployment and even came with a helpful system prompt for "Vibe coding", so I keep focused hackathon sprint.
🤖 AI-Augmented Workflow: TasksForge + Prompt Libraries + Aider
To move efficiently from planning to implementation, I followed a structured, AI-assisted development workflow inspired by this video AI Coding Tools: Prompt Libraries & Workflow Strategies for Maximum Impact!. The approach emphasizes using prompt libraries to consistently guide coding agents like Aider and Cody through a repeatable, smart dev loop.
Here’s how I applied it:
I used TasksForge.ai to generate clear, actionable user stories and requirements.
Then, using a structured prompt library, I passed those user stories into Aider chat, my AI coding assistant.
Aider chat, the best "Vibe Coding" tool ever, helped incrementally build features handling UI generation, backend integration, and logic while I focused on high-level design and creative problem solving, accelerating development without losing quality.
By combining TasksForge.ai’s structured ideation with a prompt-driven AI coding loop, I was able to go from vision to working prototype in record time.
🚀 The Final Result: Petit Prince App
You can try the live version of my app here: 👉 petitprince.val.run
Petit Prince is a thoughtful, security-aware, and mobile-friendly AI assistant designed for parents who want to share web content with their children in a safe, clear, and engaging way.
🌟 Key Features:
✅ Safety & Security First: The app doesn’t just summarize, it also filters for child-appropriate content. This ensures that inappropriate or overly complex language never reaches your child.
🧠 AI-Powered Summarization & Simplification: Parents can paste in an article URL or a block of text, then choose the target age range. The app uses Llama 4 via Cerebras API to rewrite the content in a clear, simplified tone suitable for kids.
📱 Mobile-Friendly by Design: The UI was designed to work smoothly on phones and tablets. It’s perfect for on-the-go learning moments, bedtime stories, or curious questions while commuting.
📤 Easy to Share: The app can be easily shared with other parents or family members through Facebook Messenger, allowing for collaborative parenting and meaningful family interactions.
🎯 Distraction-Free Content: One of the core ideas was to strip out casual web distractions, no ads, no cluttered layouts, no unrelated links. Just clean, focused content rewritten for kids to read, explore, and learn with curiosity.
Want to Learn More?
Check out the Petit Prince Documentation to see how the app works, the technologies involved, and the API logic behind it.
See some already generated blogs here
🎉 Conclusion
I didn’t win the hackathon, but honestly? I had an amazing time building Petit Prince, and I’m proud of what I was able to deliver in just eight hours.
Thanks to a powerful combo of:
TasksForge.ai (for scoping, planning, and rapid architecture), and the prompt library workflow featured in this video.
I submitted a fully functional project one hour before the deadline, complete with a live deployment, documentation, and a delightful concept I’m genuinely excited to keep evolving.
Whether you win or not, hackathons are about creative momentum, community, and pushing your limits. And this one definitely delivered.
Top comments (0)