Timedroid

Have you ever wanted to get insights on your employee time data. Figure out who the most outstanding employee is? Maybe analyze data? Well the answer, is here!

Comment

Inspiration

So, there I was working late as usual and wondering if the people to whom I had assigned work had done their work. One had to assume they would be right? Well, wrong! You cannot imagine the number of times at which I had realized that they weren't working but as an administrator all you can do is hope that they are and then write them their paycheck while evaluating their performance over the year. If you wanted to take things a bit further, you could fire said person but does that solve the issue? Absolutely not. It is irresponsible for us to take actions on little information and justify it on cause hence I came up with the timesheet application that tracks time and a lot more.

What it does

Get real-time insights and compare your results across the employee's time logs and figure out what has changed in their environment. Has a new employee come in that has hindered the performance of the team? Well, how do you know? Well, ITS THERE IN THE DATA! Now no one can blame you for the decisions and your decisions will be made on information that you couldn't have seen. After all, as an administrator you have a lot of things to do and too little time to do them, right? So, take the assistant's help and do miles more.

How we built it

The entire application is built using the ReactJS library and the NextJS framework leveraging a secure login logout feature using bcryptjs, storage of sessions and cookie verification using JWT, creating tables using rpc functions written using procedural postgreSQL language using supabase. Storing the NoSql queries on MongoDB and of course the hero of the show AzureOpenAIClient that performs the analysis for us. Providing us with deep insights using the GPT-4 model and presenting a chat interface for admins to chat and mine data from this application is truly your best friend.

Challenges we ran into

Some the challenges that I ran into include choosing the data structure for storage of time series data, identifying key value and unique constraints on the data, choosing between an SQL and NoSQL approach and finally preparing the data for input and providing the output of the data.

Accomplishments that we're proud of

What we learned

Working with JSON Web Tokens for establishing a session and creating cookies to store user credentials in an encrypted state, creating Middlewares that performs route protection so that only admin is privy to information of subordinates. Creating and working with the APIs and learning how to create APIs within the Next Application leverage Next API, creating an email based user verification and forgot password system for easy accessibility. And understanding the different parametes such as temperature and other arguments that went into fine tuning the model to work to our needs and finally setting up the application on vercel and hosting it. It was truly and innovative journey.

What's next for Timedroid

I plan on setting up the power of GPT4-vision for employees to upload their expense reimbursable data into the system so that the data from the expense can be studied for better and efficient management of resources. Moreover, in the next one month the application will have a portal for applying for leaves, accepting or rejecting leaves based on employee availability data. Manage shifts effectively so that employees aren't stressed out and finally for the data to be queried and represented in graphs as well.

Built With

Share this project:

Updates