Wallstreet is a technological simulation of the stock market created as the ideal platform to engage your intrest in the world of finance, Submit trades in a virtual environment before you start risking your own capital.

-Purvesh Jain


Dloud : Distributed Cloud Computing generalizes the cloud computing model to position, process, and serve data and applications from geographically distributed sites to meet requirements for performance, redundancy and regulations.

-Purvesh Jain

Facial Expression Recognition

The project is built to detect the expression of a human face first it is required to detect the different facial features such as the movements of eye, nose, lips, etc. The gray scale images of the face is used by the system to classify basic emotions.

-Yash Sonar

Carbon Footprint Calculator

This projects aims to tackle climate change and support impoverished communities across the world by calculating the carbon footprint of the individual by taking into consideration various factors.

-Tanmay Pardeshi

Covid-19 Pneumonia Detection

In light of Covid-19 taking over the world, this project has been developed to identity infected patients and uses the concepts of Transfer Learning and uses a convolutional base from a pretrained model.

-Anuraag Shankar, Asawari Walkade, Tanusri Bhowmick and Yash Sonar


This platform was developed for voting photos/sketches using QR Codes in exhibition - Picsoreel (by college student group Pictoreal). The platform used barcode based ID Cards of students to confirm identity.

-Tanmay Pardeshi and Kaustubh Odak


Chat App: This app is made using Node.js in the backend and basic HTML, CSS and JS in the front end. The app also uses Express.js to create API endpoints that can be accessed by users.

-Yash Sonar


Wallstreet is a stock market simulator, where the players get an opportunity to earn truckloads of virtual money. All that the players have to do is, follow the news, select the stock and wait for the price to skyrocket. On the flip side, they learn how to recover losses from stocks that have tanked. The biggest takeaway from the Wallstreet is that the inexperienced players leave with a huge understanding about the stock market and are ready to step out into the street to make some real money! This platform made it's debut at PISB's anual technical festival, the game successfully handled 1000+ players flawlessly.

Git Hub


Dloud - The Distributed Cloud, here we tried to create our own cloud storage system in a Distributed way. The main aim of our project- Distributed Cloud is to achieve higher speeds as compared to standard FTP server and to utilize free space on all desktops in a certain lab. The ‘Dloud’ contains multiple paths for files to be uploaded to data servers as well as makes way for faster retrieval of huge files like movies. Due to multi-threading, we achieve high speed as compared to the single link FTP servers. This helps in utilizing free space available of huge number of computers in the campus in a secure way as the data is split up and distributed on the network.

Git Hub


This model carried out the task of Facial Expression Recognition using a Convolutional Neural Network (CNN). This CNN was implemented using Keras backed by Tensorflow. This study proposes a learning emotion recognition model, which consists of three stages: Feature extraction, subset feature and emotion classifier. Around a 1000 images belonging to each expression category were used to train the CNN. The CNN is hosted online by using Flask which is a backend framework in Python. Users can see the real time implementation using the front camera of their laptops.


A carbon footprint is defined as the total amount of greenhouse gases produced to directly and indirectly support human activities, usually expressed in equivalent tons of carbon dioxide (CO2).The carbon footprint is a very powerful tool to understand the impact of personal behaviour on global warming. Most people are shocked when they see the amount of CO2 their activities create! If you personally want to contribute to stop global warming, the calculation and constant monitoring of your personal carbon footprint is essential. This project is a simple web app to calculate carbon footprint emissions of a user in day to day life. With the help of Django to create the backend of the web server, SQLite for the database, while simultaneously integrating the website's front end using HTML,CSS and Javascript. An API of the maharashtra state electric board is used for consumer details.

Git Hub


This model is a Convolutional Neural Network that uses the concepts of Transfer Learning. It uses a Convolution base of a pretrained model that helped users to detect which X-ray images have Pneumonia and which don’t. Transfer Learning was necessary to be used because of the limited number of images available to train the model. This model achieved a whopping accuracy of 99% with a 5-fold cross validation on the total dataset conatining 591 images. This application is very trustworthy for implementation in the real world and is ready for deployment.

Git Hub


A working project used as an online voting platform with a QR Code and Barcode scanner.
Authentication :
Volunteers register the audience from admin login (/adminlogin) using barcode scanner. Once registered, the audience are able to login from the home page(/) by scanning their ID Card barcode. This is done to ensure that no one votes twice.
Voting :
Once logged in, audience need to scan the QR Code of the photo/sketch they want to vote. (/vote). Once the audience has finished voting, their votes are submitted for calculation.
Results :
The results are calculated at runtime at route/count (accessible only to the admin and volunteers).

Git Hub


This app is made using Node.js in the backend and basic HTML, CSS and JS in the front end. The app also uses Express.js to create API endpoints that can be accessed by users. The app is hosted online on Heroku. Users can join rooms to chat with each other. The communication of the messages is done using which can be used with Node.js to send (emit) and receive messages. There is no database used in this for data privacy and the data is deleted as soon as the user leaves the chat room.

Git Hub  See Project


Covid-19 Global Hackathon

Covid-19 Global Hackathon organised by DevPost brought about thousands of students from all over the world to compete and to use technology to bring about effective changes to help curb the pandemic.

-Aditya Vadalkar, Tanuj Agrawal and Tanmay Jain

Pune Open Data Hackathon

The open data hackathon was conceptualized as an event where problems & challenges faced by the citizens are presented to the participants. The hackathon draws in the most skilled technological students invested in different domains.

-Yash Sonar

Pune Open Data Hackathon

Cleaning and Analysis of New Vehicles Registrations in Maharashtra. Contents include cleaned data and visualisations of various regions' RTO vehicle registrations in Maharashtra through years 2000-2017.

-Aboli Marathe, Hussain Naharpurawala, Mufaddal Diwan and Saarth Deshpande


The proposers of selected ideas prove their Ideas by building a prototype for establishing the Proof of Concept and are required to demonstrate their prototypes and make presentations to contribute to a noble cause.

-Harsh Sakhrani, Kaustubh Odak, Saloni Parekh, Snigdha Singh and Tanmay Pardeshi


What it does:
Corona Spy is a smartphone app aimed at tracking the spread of COVID-19 among its users by forming links among them as they come in close contact with each other. WIth the forming of such links, a web structure gets formed which links users infected by the virus to those who are at maximum risk of getting infected due to contact. This structure extends furthermore to track users who may need to be tested or quarantined in the near future. The objective of the web structure is to make tracking of potential victims of COVID-19 streamlined and efficient which will help reduce the R0(R-Knaught) value of the disease significantly.

How it does it:
Corona-Spy makes use of the BLE(Bluetooth Low Energy) beacon technology of a smartphone to detect other BLE beacon equipped smartphones in a defined range of area around it. The range of area is defined by setting an RSSI threshold to limit the range of Bluetooth signals fetched by the beacon. This will ensure that users who are not in the close vicinity of a particular user remain undetected. To form a link between two users, a unique ID of the two users is exchanged via the BLE Beacon and gets stored in each other’s databases along with the time, date and geolocation information of the location where the contact occurred. Every user of the app has his/her unique ID associated with their account, which will help identify them during the tracking process. The UUID of every app user’s beacon will be the same to distinguish the BLE signals received by the users of the app from any stray bluetooth signals.

Git Hub Git Hub Git Hub See Project


The hackathon is conducted like a laboratory session where the students used methods and techniques to extract knowledge and insights from the data in various forms. They work towards developing solutions for the problem statements. Experienced professionals were available to guide the participants during the session – to generate ideas and steer them towards the right solution approach. We prepared a geospatial analysis of taxation rates and property rates information across Pune. This analysis provided clear understanding of the data which otherwise was difficult to comprehend by normal individuals.


Contents include cleaned data and visualisations of various regions' RTO vehicle registrations in Maharashtra through years 2000-2017.
1. Create a secondary dataset containing all years' data combined together.
2. Plot a graph of Population Density versus New Registrations for the year 2011.
3. Plot various region-wise graphs for change in number of various types of vehicles such as two wheelers, cars, commercial vehicles etc. over the given period of 17 years. (60+ graphs have been plotted)
4. Plot various region-wise pie charts comparing change in number of private / commercial vehicles over the given period of 17 years. (15+ pie charts have been plotted)
5. Create an interactive Webpage using Mapping for easier access to the visualized data {A map of Maharashtra with boundaries of various regions, and on clicking each region, one is able to see numerous visualizations pertaining to that region.}
6. Write a Python code to visualize CSV / TXT file and save the visualized line graph in .jpeg or .png format.

Git Hub


CLASP - COVID-19 Literature Analysis and Summarization Platform. Backend implemented using flask and frontend is in React. REST APIs are also used in this project. Features such as Text Summarization, Information Retrieval, Question and Answers have been implemented. New features such as real time news and anomaly detection is going to be used for the final project. Web deployment using Nginx, Gunicorn has been done on Microsoft Azure VM. Users can filter documents and search for keywords or similar images with the help of this platfrom.

See Project