Artificial intelligence is powerful and flawed, and I enjoy finding ways to make it better by optimizing what machines and humans do best. Someday machines will be nearly flawless at processing knowledge in ways that will make our lives better, but we are not there yet. This app trains humans to speak to machines in almost natural language, with the added bonus of reminding users that there are humans on the other side of the technology that they are using. It is a more efficient, conversational search tool and a web-based predecessor to the human API.

I built it using Twitter data for content, MaidSafe for secure, encrypted storage, and standard front-end bootstrap with HTML, CSS, and Javascript.

It was challenging because I needed more experience with bots before trying to use them in a sprint. For now, bots live on the web, and I was trying to force them into a real-world emergency scenario too soon.

I am proud that I was able to string together more "natural" tools while suggesting a platform that could still accelerate the ability of machines to understand humans. I also learned that even at hackathons, I should start with my passion - in this case data ownership and positive user experiences - and only focus on the APIs if they really enhance what I can build during a sprint.

I plan to continue working on the UI, using data that comes from user choices - particularly search - to deliver insights to individuals and training data to machines.

Built With

Share this project: