Hi, my name is Shamik Bera
I'm a Full Stack Developer, ML, Ai and a CyberSec Enthusiast.

Smart Brain is one of the current projects I have been working on,It consist of a Dashboard and User needs to Login/Register to get there and the user needs to upload a image wih a face in the image, Using the Face Recognition Model from Clarifai API,the model detects and highlights the face region with a box around it, and also the user recieves some points for doing so.It is made using React,Node JS,Express and PostgreSQL as database to store user and points awarded.

A Real-time Chat app that allows to have a group chat with your friends or any other person verified using it.It is made using React as framework and for backend, I used Firebase with Firestone as the database here.

The Cyber Experts and Police have been fighting cyber crimes since the invention of the computer.So we wanted to help our digital heroes with a efficient management system to record and store cyber records with the information and also an effective reporting system to report the same to the victims as well as police officers.The Front End uses EJS and for Back End, it uses Node JS along with libraries like mysql,bycrypt.js for connection to database and security of website to prevent SQL and XSS injection attacks.

Haydio is gesture controlled radio web app which detects hand gestures to turn on and off the radio whenever needed. I am currently working on improving the model made using Deep Neural Networks which is provided by TensorFlow with some hyperparameter tuning and such.Later on, I will also come up with features like volume up/down changing the radio station and such.It is made using custom HTML,CSS and Vanilla Javascript and for the Machine Learning Model, I am using the Handtrack.js library by TensorFlow which allows me to use the model for Hand Recognition.

The Movie Recommendation System project aims to recommend 5 movies based on what movie the user had watched before.It uses Content-Based Recommendation Architecture where I used a movie dataset with names,genres,tags and such and then using Bag of Words technique, I derived cosine distance between the searched movie content which contained all the info from overview to keywords and even actors with all the other movie content present in the dataset and as an output, it gives 5 such movies with least cosine distance from all the content vectors present in the dataset.It was made using sklearn for modules like cosine similarity,Count Vectorizer and such and then made a simple web UI using Streamlit.