Dashboard with AI recommendation and analysis tools
Entry point for environment location

Spacium.AI

Features

Real-time collection of sensor data (temperature, humidity) + filler data
Automatic calculation of environmental scores (0–100)
Safe thresholds detection and alert system
JSON output for easy integration with back-end systems
Handles missing or extra sensor fields gracefully
Supports multiple readings for trend analysis

Inspiration

Safety and professionalism are values that deeply align with our group's morals. Having an ideal direction of what to build before the start of the event, narrowing it down with the company challenges gave us the foundation for a solid project.

What it does

This project is an environmental monitoring and analysis tool for surgery rooms. It collects sensor data such as temperature, humidity, CO2, PM2.5, TVOC, pressure, and light, then:

Compares readings against medical/surgical standards
Calculates sterility, storage, and compliance scores
Flags unsafe conditions with alerts
Provides actionable recommendations for maintaining safe environmental conditions
Outputs results in machine-readable JSON suitable for dashboards, AI analysis, or alerting systems

How we built it

Backend: Python / FastAPI
Database: Supabase for storing sensor readings and AI analysis
AI Analysis: Claude for automated evaluation and recommendations
Frontend: React Dashboard for visualization (React)

Challenges we ran into

Sensor setup, AI/API integrations, http requests between front and back-end

Accomplishments that we're proud of

Working with real sensor data for accurate analysis in different room types, with recommendation systems and alerts through AI integration

What we learned

Through this project, the team was able to refine their program architecture skills. It was a great the team's first hackathon and presented plenty of opportunities to depthen our skills and develop across multiple fields

What's next for Spacium.AI

Spacium.AI has proven to be a promising and powerful tool for management of different environment types. It has the potential to be improved on through broader sensor and situation integration. It can be utilized across multiple fields in professional settings and personal home monitoring.

Setup

This project relies on the use of a SHT45 Adafruit Trinkey sensor for the collection of temperature and humidity data. Other dependencies for the use of the sensor are Thonny IDE and PythonCircuit. Supabase was used for the SQL database and FastAPI was used alongside Uvicorn for http requests. Pydantic for data validation and Anthropic for AI API calls.