Amanpreet Singh Saimbhi

Diligent and meticulous individual pursuing B.Tech in computer science. I am inclined towards challenging projects having constrained timelines. At the core, I am a self learner and a team player who’s open to exploring more in the discipline of software development. Also, I am extremely passionate about theatre. I've done various courses like Core Java, Python Specialisation, Django 2.1, Machine Learning(Andrew Ng), Deep Learning Specialisation(DeepLearning.ai) to keep up with the latest technology. I've a strong zeal to learn new skills. Only learning new skills itself is not enough to make you a better programmer, that is why I try to implement the skills which I learn on to projects. My project work includes : Library management system using core java : This is a rather simple project, in which the admin can view, issue or return the books using the GUI and the required changes will be made to the database,
NASA's Dataset Analysis and Visualisation : Using NASA's Dataset API, loads the JSON file into the cache memory, and then stores the raw data in the database. Then cleaning and modelling of data is done. Then data is used for Analysis and Visualisation using Pandas , Numpy and Matplotlib,

Portfolio Website : A Django website which is basically my portfolio, you can check my work, resume, LinkedIn and blogs in it. It is deployed using DigitalOcean, Nginx , Gunicorn,

Product Hunt Clone : Product hunt is a website that lets users share and discover new products. I have made a clone of this website using Django 2.1 where users can Signup, Login and Logout with an authentication system. Users can create, view and upvote products,

Image Classification using CNN : Web app inspired from the hotdog app in the ''Silicon Valley'' show. Model is a fine tuned version of VGG-16, which is deployed as a web service using flask. Decided to make use of transfer learning to fine tune the model for my classification task.

Writing Like Shakespeare using LSTM : A model to generate Shakespeare poems on top of the first line given by the user. learning from a dataset of collection of Shakespearian poems. Using LSTM cells, learned longer term dependencies that span many characters in the text--e.g., where a character appearing somewhere a sequence can influence what should be a different character much much later in the sequence.
Also, some other projects of deep CNN's like visual object detection(with 80 classes), Neural style transfer( in which a content image and a style image are used to synthesise a new image), 1:K Face Recognition(which solves the one-shot learning problem).