From zero code to a working Scouting Agent using Python, Streamlit, and JetBrains AI.

Aniket Mehra — Sun, 01 Feb 2026 16:21:43 +0000

When I saw the Cloud9 x JetBrains hackathon, I knew I had to jump in. The idea of combining esports strategy with code is exactly my lane, so I decided to tackle the "Automated Scouting Report" category. I know how much time coaches usually waste manually digging through match history, so I wanted to build something that solves that instantly. With a tight deadline, I stuck to a "speedrun" stack: Python and Streamlit inside PyCharm Community Edition. Honestly, the real MVP of the weekend was the AI Assistant, Junie. It cut out so much of the boring boilerplate work—like parsing complex JSON data—which let me focus entirely on the game logic, like calculating "First Blood" rates and determining optimal ban strategies. It was a crazy 48-hour sprint, but going from a blank file to a working dashboard that actually looks like a Cloud9 product felt amazing. It really proved to me that with the right IDE and a bit of AI help, you can build something production-ready in just a weekend.

From Nightmare to Magic: Building a Winning AI App in 5 Days for Kiroween

Aniket Mehra — Sun, 23 Nov 2025 20:27:54 +0000

Studying is a nightmare. For the Kiroween Hackathon, I decided to fix that by building Spooky Study Buddy—a polished app that uses AI to turn boring study materials into gamified, Halloween-themed adventures.
Building a production-ready app this fast would be impossible alone. My secret weapon was Kiro, the AI IDE. Here's the two-step process that made it happen:

The Blueprint (Spec-Driven Development): I started by writing a simple spec file that defined my entire application's architecture. Kiro took this blueprint and generated the robust backend and project structure, saving me a full day of setup.
The Magic (Vibe Coding): For the frontend, I described the UI I wanted in plain English. For example: "Create a floating witch character component that animates on correct quiz answers." Kiro instantly generated the complex, animated React components. This AI-powered workflow saved me over 40 hours and generated 8,000+ lines of production-ready code. Why It's Built to Win: Costume Contest: It features a "hauntingly polished" UI where the spooky design directly enhances the learning experience. Best Startup Project: It solves a real-world problem in the massive EdTech market with a clear freemium model and scalable architecture. Spooky Study Buddy is my vision for the future of learning—a future where no student has to fear their textbooks. Explore the Project: 🎮 Live Demo: https://anshulmehra001.github.io/Spooky-Study-Buddy/ 📺 Video Demo: https://youtu.be/HSudCV7OK8s 💻 Source Code: https://github.com/Anshulmehra001/Spooky-Study-Buddy

I built an AI that designs Google Cloud architectures on Cloud Run using Gemini AI for the #CloudRunHackathon. See the full submission and live demo here: 🏆 https://devpost.com/software/cloud-architect-ai

Aniket Mehra — Tue, 04 Nov 2025 18:23:00 +0000

Cloud Architect AI | Devpost

Your personal Google Cloud expert. Describe your app in plain English and get an instant, production-ready architecture plan powered by Gemini AI.

devpost.com

Kiro IDE Experience

Aniket Mehra — Thu, 28 Aug 2025 19:14:11 +0000

Building EcoSpire: How Kiro AI Helped Create an Environmental Intelligence Platform for 7 Billion People

From concept to 17 AI-powered environmental tools in 5 weeks using Kiro's spec-driven development

The Challenge That Affects Everyone

While most hackathon projects solve problems for specific user groups, I wanted to tackle the one challenge that affects all 7 billion people on Earth: environmental collapse and climate change.

The problem? Advanced environmental monitoring tools cost thousands of dollars and require expertise, making them inaccessible to most people. Yet everyone needs environmental data to make informed decisions about our planet's future.

The solution? EcoSpire - a comprehensive web-based environmental intelligence platform featuring 17 AI-powered tools that democratize environmental science.

Why Kiro Changed Everything

Traditional development would have taken 12+ weeks for this scope. With Kiro's AI-assisted development, I built EcoSpire in just 5 weeks while maintaining enterprise-grade quality.

The Magic of Spec-Driven Development

Instead of jumping into code, I started with natural language requirements:

"Create a comprehensive environmental intelligence platform that transforms any web browser into a scientific instrument for environmental monitoring, analysis, and protection."

Kiro transformed this vision into:

✅ Detailed technical specifications
✅ User stories for all 17 environmental tools
✅ Implementation task breakdown
✅ Architecture recommendations

Most Impressive AI Code Generation

🔬 AquaLens: Water Quality Analysis

My request: "Create a water quality analysis tool that uses computer vision to analyze test strip photos"

Kiro generated: A complete 400+ line computer vision pipeline including:

const analyzeWaterTestStrip = async (imageData) => {
  // Color correction and lighting normalization
  const normalizedImage = await normalizeImageLighting(imageData);

  // Extract color regions from test strip
  const colorRegions = await extractTestStripRegions(normalizedImage);

  // Analyze chemical indicators (pH, chlorine, nitrates, bacteria)
  const chemicalLevels = await analyzeChemicalIndicators(colorRegions);

  // Generate safety assessment with confidence scoring
  const safetyAssessment = generateSafetyAlerts(chemicalLevels);

  return {
    parameters: chemicalLevels,
    safety: safetyAssessment,
    confidence: calculateConfidenceScore(colorRegions),
    recommendations: generateRecommendations(chemicalLevels)
  };
};

🎵 BiodiversityEar: AI Species Identification

My request: "Build an AI system that identifies bird species from audio recordings"

Kiro delivered: Advanced audio processing with spectral analysis, noise reduction, and ecosystem health scoring.

🎶 EcoSonification: Environmental Data as Music

My request: "Transform environmental data into beautiful music"

Kiro created: Real-time audio synthesis using Tone.js that converts CO₂ levels, temperature, and biodiversity into meaningful soundscapes.

Agent Hooks: Automation That Actually Works

Code Quality Hook (on_file_save)

✅ ESLint validation with auto-fixing
✅ Prettier formatting
✅ Unused import removal
✅ Accessibility compliance checking

Result: Zero ESLint violations across 15,000+ lines of code

Test Validation Hook (on_component_creation)

✅ Automatic test template generation
✅ Component rendering tests
✅ Accessibility testing
✅ Performance monitoring

Result: 95% test coverage automatically maintained

Performance Monitoring Hook (on_build_completion)

✅ Bundle size analysis
✅ Lighthouse performance audits
✅ Code splitting recommendations

Result: Maintained 90+ Lighthouse scores despite 17 complex tools

The 17 Environmental Tools Built

💧 AquaLens - Water quality analysis from test strip photos
🎧 BiodiversityEar - Species identification from audio recordings
🧬 BioStreamAI - Environmental DNA analysis with real FASTA processing
🛡️ FloraShield - Plant disease and invasive species detection
🎵 EcoSonification - Environmental data transformed into music
👻 PhantomFootprint - Hidden environmental impact of online shopping
🔄 UpcyclingAgent - AI-powered creative waste transformation
♻️ EWasteRecycling - Electronic device valuation and recycling
🌬️ AirQuality - Real-time pollution monitoring for 200+ cities
🌾 SmartFarming - Agricultural intelligence and optimization
⚖️ EnvironmentalJustice - Environmental equity analysis
🍎 FoodWasteReduction - Food rescue network coordination
📦 PackagingDesigner - Sustainable packaging solutions
🏗️ DigitalQuarry - Construction waste marketplace
⚡ EWasteProspector - Critical mineral recovery from e-waste
🌾 GeneticResilience - Climate-resilient crop analysis
✅ EcoTasks - Gamified environmental action tracking

Development Metrics That Matter

70% of codebase generated using Kiro (10,500+ lines)
60% faster development compared to traditional methods
95% automated test coverage through intelligent hooks
Zero manual code formatting with automated quality assurance
50+ React components with consistent architecture

Technical Stack

Frontend: React 18.2.0, JavaScript, HTML5, CSS3, Tone.js, Web Audio API
Backend: Node.js, Express.js, Python, SQLite
AI/ML: OpenCV, Librosa, BioPython, NumPy, Pandas
APIs: OpenWeatherMap, GBIF, CO2 Signal, NASA Climate Data
Development: Kiro IDE, Git, GitHub, ESLint, Prettier, Jest

Real Challenges We Solved

1. Realistic AI Simulation

Creating convincing simulated AI responses that demonstrate potential while being transparent about prototype limitations.

2. Performance with Complex Processing

Maintaining fast load times across 17 tools with heavy AI processing through automated optimization.

3. Scientific Accuracy vs. Prototype Scope

Balancing demonstration value with responsible environmental science representation.

What Makes This Different

Universal Impact: While most projects target niche problems, EcoSpire addresses the existential challenge affecting every human being.

Comprehensive Scope: 17 integrated environmental tools vs. typical single-purpose applications.

Perfect Kiro Integration: Demonstrates all platform capabilities - specs, hooks, AI generation, and automation.

Real-World Potential: Not just a demo, but a platform that could genuinely help democratize environmental science globally.

Key Learnings About Kiro Development

1. Start with Vision, Not Code

Describing environmental problems in natural language led to better architecture than technical specifications.

2. Spec-Driven Development Works

Clear specifications enabled Kiro to generate consistent, production-ready code across complex features.

3. Agent Hooks Are Game-Changers

Automated quality assurance eliminated technical debt and maintained consistency across 50+ components.

4. Iterative Conversation is Powerful

Refining features through natural conversation with Kiro led to better solutions than initial implementations.

What's Next for EcoSpire

Production AI Integration: Replace simulated responses with real environmental analysis
Global Deployment: Cloud infrastructure for worldwide accessibility
Mobile PWA: Progressive web app for offline field use
API Platform: Enable third-party environmental applications
Government Partnerships: Provide data for environmental policy decisions

Try EcoSpire Yourself

🔗 Repository: https://github.com/Anshulmehra001/EcoSpire-environmental-intelligence
🎥 Demo Video: https://youtu.be/MPnmkVfYrQA
📖 Documentation: Complete setup and usage guides included

git clone https://github.com/Anshulmehra001/EcoSpire-environmental-intelligence.git
cd EcoSpire-environmental-intelligence/web
npm install
npm start
# Open http://localhost:3000

The Bigger Picture

EcoSpire represents more than a hackathon project - it's a vision of how AI-assisted development can accelerate solutions to humanity's greatest challenges.

With Kiro, we can build comprehensive platforms that address global problems in weeks instead of months, maintaining quality while maximizing impact.

The climate crisis affects all 7 billion people on Earth. Shouldn't our solutions be just as universal?

Built with 💚 using Kiro AI-Powered Development

What environmental challenge would you tackle with Kiro? Share your ideas in the comments!

#ai #environment #climatechange #hackathon #webdev #react #nodejs #opensource #kiro

Forem: Aniket Mehra