Inspiration

An old laptop, clothes that no longer fit, or a drawer full of batteries can leave you with the same question: what should I do with these?

I wanted to make that decision easier. E-COnnect helps people find practical ways to keep useful items in circulation and understand their local disposal options.

The problem—and how I know it’s real

I started with published recycling guidance for Aurora and nearby communities in Illinois. The rules show why finding a recycling location isn’t always enough.

For example, Naperville’s electronics collection excludes loose batteries. Its household hazardous-waste program accepts batteries and oil-based paint but excludes latex paint. Both programs operate at the same address.

These differences shaped the project: people need guidance based on the specific item and its condition. This first version draws on public sources; interviews with residents and collection staff are a next step.

What it does

Users can search for an everyday item or try identifying it from a photo. After confirming the item and its condition, they receive suggested next steps for repair, reuse, donation, recycling, or special handling.

Local listings include acceptance notes and links to official sources. Users can save preparation checklists, mark actions complete, and export a copy.

The current directory focuses on Aurora, Naperville, and the Fox Valley. Photos are processed in the browser, and no account is required.

How I built it

I built E-COnnect with HTML, CSS, and JavaScript, with AI assistance during development. TensorFlow.js and COCO-SSD provide browser-based image suggestions.

A separate rules engine matches items and conditions to curated guidance. Checklists use browser local storage, and GitHub Pages hosts the website. The public repository includes the source code, documentation, and eleven passing logic tests.

Challenges I ran into

One challenge was handling exceptions. A working laptop and a device with a swollen battery should not receive the same recommendation.

Another was image recognition. Recognizing an object does not establish its material, condition, or safety. I addressed this by requiring confirmation and keeping manual selection available.

Accomplishments I’m proud of

I’m proud of connecting information to an action someone can follow through on. E-COnnect brings item guidance, local resources, and a saved checklist into one working app.

It also keeps its impact claims honest: completed actions are self-reported, not measured carbon savings.

What I learned

I learned that useful sustainability tools depend on local details and clear explanations. Handling uncertainty well matters just as much as recognizing an item correctly.

What’s next

I want to test E-COnnect with residents and collection staff, improve the guidance using their feedback, and add more verified communities. I would also like to introduce translations and a clearer process for keeping local information current.

Built With

Share this project:

Updates

Submission history