Watchdog

Inspiration

Inspired by NextEra Energy's challenge at Knight Hacks 2026, we developed Watchdog, a platform that transforms synthetic Boston Dynamics SPOT robot inspection data into actionable insights.

Our goal was simple: A completed inspection should never mean an overlooked problem.

What it does

Watchdog helps operators monitor robotic inspections through an interactive dashboard. Users can replay missions, follow SPOT's movements, identify hidden equipment anomalies, and prioritize maintenance decisions.

Our platform also provides equipment histories, performance analytics, interactive facility maps, and simulated scenarios involving extreme weather and obstacles.

How we built it

We built Watchdog using Next.js, React, TypeScript, and Tailwind CSS, with the interface designed in Figma.

Our custom simulation engine uses A pathfinding and Canvas 2D* to animate SPOT navigating inspection routes.

The backend uses Node.js, Express, and Neon PostgreSQL to manage inspection records, sensor readings, anomalies, and operator decisions.

We also explored AI integration using a Hugging Face summarization model, allowing users to generate draft mission summaries with validation checks against inspection records.

The application is deployed using Vercel and Railway.

Challenges we ran into

Our first challenge was defining team responsibilities, designing the architecture, and coordinating development. Some team members were unfamiliar with GitHub workflows, adding to the learning curve.

We also faced challenges integrating the frontend, backend, and simulation engine within the hackathon's limited timeframe.

Another challenge was implementing an open-source AI model. It sometimes generated information that wasn't supported by our inspection data, teaching us the importance of validating AI-generated responses.

Accomplishments that we're proud of

We're proud of transforming our initial idea into a deployed, full-stack application in one weekend.

We successfully integrated a custom robotic simulation, interactive dashboard, cloud database, mission analytics, and operator decision tracking.

Most importantly, we created a system that demonstrates how robotic inspection data can help operators recognize problems that might otherwise go unnoticed.

What we learned

We strengthened our skills in GitHub collaboration, full-stack development, relational databases, API integration, simulation, and AI experimentation.

We also learned the importance of planning, defining responsibilities, communicating across different development areas, and managing our time effectively.

Above all, we learned that building a successful application requires both technical skills and teamwork.

What's next for Watchdog

We hope to connect Watchdog to live robotic inspection systems, improve the accuracy of AI-generated insights, and expand predictive maintenance capabilities.

Our vision is to help energy operators identify problems earlier, improve inspection efficiency, and strengthen infrastructure reliability.