CardCat

Inspiration

Having gone to many conferences in the past, I've ended up with a stack of business cards that lay unused due to the high time cost of going through them. My goal with CardCat was to lower that cost, making networking easier.

What it does

It takes in a image via a MMS message, parses text & scans for contact info in that text, then returns a VCard of that contact information.

How I built it

The entire workflow is in python, at first it was run locally via a Flask server, then once I had a reasonably stable version I migrated to Google Cloud functions to allow for lower latency & easier scaling.

Challenges I ran into

Windows. Line. Endings.

Accomplishments that I'm proud of

Creating and shipping this project all in one day, I was skeptical I would finish at first.

What I learned

Using APIs is an amazing way to get a lot done in a little time, why solve a problem twice?

What's next for CardCat

It would be great to integrate it into a mobile application that allows people to correct errors that the parser made, and storing these corrections to improve predictions in the future.

Try it out (It's not a url, so I can't put this below)

Text a picture of a business card to (609) 785-3037 & it will reply with a VCard. (Note: it can take up to a minute to reply)

Built With

• google-cloud
• google-cloud-language
• google-cloud-vision
• google-could-functions
• python
• twilio