BusiScan - AngelHack '15

AngelHack is not for the faint of heart. It lasts 24 hours, has a lot of sponsor backing and a ridiculous amount of Red Bull. Despite this being my second year at this, it turned out to be just as exhausting and exciting as the last.

I went in looking for a team to build with and ended up with some folks who really knew their backend/web scripts well and could back me up on the Android side of things.

Our idea was to make a business card scanner for SMBs. It would be free for scanning cards and adding them to your contacts - which is a good achievement in itself, considering most current apps are paid. Where we would be able to make a difference is that it would tell the business where their cards were being scanned and provide more analytics around the paradigm.

I built the Android app for the front end of this service. The input could be a stored picture or one taken from within the app. The app called on the backend we built and served up JSON for the app to consume. We managed to pull name, company, phone number and email addresses from most regular-looking business cards with good accuracy. Under the hood, we used sponsor HP's IdolOnDemand API for entity extraction and OCR document scanning.

Shoe Gamble - Best Smartphone app at the HH Brown Shoe Hackathon

The mandate - "hack shoes" in 8 hours. The tool - an API that had the access to the full range of shoes under the HH Brown brand. 

What I developed was an Android app called Shoe Gamble. The idea of the app was simple enough - take in some basic information from the user, such as their gender and the minimum and maximum that they were willing to spend on a pair of shoes. The app would generate a random shoe that would give the user the option to buy that pair of shoes at a discounted price. If the user didn't like that particular pair of shoes, they could have another go. However, the percent discount offered this time would be lower. If the user wanted a third go at it, they would need to wait 24 hours.

This mechanic did what many apps struggle to - make customers keep coming back. If the company was inclined to, they could also game the "randomness" of the generated shoe. It would be easy enough to raise the odds of a particular pair of shoes coming up to cut down on excess stock, for example.

My app was chosen as the best one at the event for the idea, and for the complete implementation as a native Android app. Being the sole developer on my team also let me hold onto all the winnings!