The inspiration came to us from a plugin called Insights which is integrated in Microsoft Outlook. Insights is platform specific and does not help remote workers with other parts of their work. We wanted to build an AI based tool that provides insights on tasks at hand whether it's Gmail, Slack , Jira or Zoom meeting. Remote workers are tired of constant app switching and they deserve better. HiGrow AI is here to rescue remote workers all across the globe.
What it does
HiGrow is AI based platform for task insights. It gathers data from different sources such as Jira, Gmail and Slack for analytics. Allows individual remote workers to use high level overview and keep track of work.Cloud based dashboard is AI enabled which provides suggestions to remote workers. Automate repetitive tasks on user permission. Help employees analyze tasks and take clear actions using Machine Learning. Answer remote workers concerns using Natural Language Processing.
How we built it
In the time available, we were only able to build a mockup which aims to show what should be possible. Our investigations of slack/Jira APIs and other solution space aspects provided useful experience. Implementing the AI parts in the time was simply not possible. The following paragraphs describe the envisaged work rather than what we have built.
We built the server using a Python library Pytorch for Artificial Intelligence. AWS to host Web Platform. We've also built a web app using Node.JS deployed via Netlify as a single page application which communicated with our database. Check out our Cloud app in the links below.
Challenges we ran into
Artificial Intelligence framework. Real Time data visualization. Generating useful high level metrics for the large amounts of data our platform collects. Machine Learning and Natural Language Processing framework
Accomplishments that we're proud of
We've built a mock bot to demonstrate in-app learning. Legal aspect were covered to protect user privacy and data.
What we learned
On the engineering side, we had to learn how to develop an AI Bot, Perform Machine Learning , Perform Natural Language Processing on user request
In general, we've learned that there is too much distraction in a remote worker’s life and not enough good tools. No high level overview from various apps that should provide analytics for improvement. Remote workers get frustrated on critical decisions due to lack of clarity.
What's next for HiGrow AI
Short term : Train Engines and develop custom AI framework for various job roles and skill sets. Add more applications to the platform
Long term: Distribute the platform services globally keeping regulations in mind.