🚀 Building RoadWatch using Kiro IDE

Github Repo Link - https://github.com/eklavyadev/roadwatch-kiro

Team Introduction 🚦

We are Team Civic Forge, a team of developers who enjoy building practical solutions for real‑world problems.

Team Members:

• Eklavya Chandra

• Abhijeet Mishra

• Aditya Mishra

• Aayush Shukla

At the HackXIOS hackathon, we worked on RoadWatch, a civic‑tech platform focused on improving how road damage and potholes are reported and visualized.

---

What We Built: RoadWatch 🛣️

RoadWatch is a validation‑driven pothole reporting platform that allows citizens to report road damage using photo evidence and GPS‑verified locations.

Instead of collecting large volumes of unverified complaints, RoadWatch focuses on data quality and reliability. The system ensures:

• Mandatory image uploads

• GPS accuracy checks before submission

• Backend duplicate prevention within a defined radius

• Clear validation errors for users

This helps ensure that reported issues are meaningful, non‑redundant, and easier to evaluate.

---

Tools & Technologies We Used 🛠️

• Next.js for frontend and API routes

• Supabase for database and storage

• Google Maps API for location visualization

• AI service (deployed on Render) for image‑based validation after local testing

• Vercel for continuous deployment

• Kiro IDE for structured development, planning, and workflow assistance

---

How We Used Kiro IDE ✨

Kiro IDE played an important role in keeping our development organized and structured, especially under hackathon time pressure.

We used Kiro IDE for:

• Planning system architecture and workflows

• Refining existing features and improving usability

• Maintaining a disciplined development process

One example was the suggestion to consolidate all reported potholes into a single interactive map, which significantly improved visibility and evaluation for judges.

---

Development Workflow & Commit Discipline 🔁

One challenge we consciously addressed was maintaining a clean and review‑friendly commit history, since hackathon judges often look at Git activity.

To solve this, we followed a strict workflow:

• Code was committed after every small logical change (roughly every ~20 lines of code) [automated using Kiro AI]

• After every 5 commits, the code was pushed and deployed to Vercel

• This ensured steady progress, transparency, and no large code dumps

Kiro IDE helped reinforce this discipline by keeping tasks well‑structured and changes incremental.

---

AI Integration & Impact Page 🤖

For the AI component, we:

• Tested the service locally

• Deployed it on Render for stability and easy access

• Built a dedicated Impact page to help judges quickly evaluate outputs and understand the system’s effect

This made evaluation smoother and reduced the need for manual explanation during demos.

---

Validation System (Core Focus) ✅

A major part of RoadWatch is its validation system, which ensures data integrity:

• Image size and type validation

• Mandatory field checks

• GPS accuracy enforcement

• Backend duplicate detection for nearby reports

This validation‑first approach helped us prioritize accuracy over volume, which aligns well with real‑world civic use cases.

---

Overall HackXIOS Experience 🌟

HackXIOS was a great experience that emphasized process, clarity, and tooling, not just final output.

The combination of:

• disciplined commits

• continuous deployments

• validation‑first design

• and structured tool usage

helped us deliver a project that was not only functional, but also easy to review and evaluate.

---

Final Thoughts

Building RoadWatch taught us the importance of combining good engineering practices with the right tools.
Kiro IDE helped us stay organized, improve features, and maintain clarity, while the hackathon environment pushed us to be disciplined and thoughtful in our approach.

Overall, HackXIOS was a rewarding and educational experience 🚀