Writing & Sessions
Articles and resources I’ve published, plus recordings of talks and workshops I’ve delivered on AI, ML, and building software.
Articles
Sessions
Sep 2023
The Gandalf Model is a deep learning solution designed to protect young web users from harmful online content. Using DistilBERT and BiLSTM, it classifies websites into ten categories, including NSFW, with 80.1% accuracy. Trained on over 400,000 labeled entries, the model outperforms traditional approaches and can be integrated with a proxy server to block access to undesirable sites, enabling safer internet use.
Jan 2023
This demo highlights an innovative way to communicate without the internet by using light signals. Watch how data is transmitted seamlessly through light, enabling offline, wireless communication.
May 2022
Kaggle is an online community platform for data scientists and machine learning enthusiasts. Kaggle allows users to work together on data science problems, locate and publish datasets, use GPU integrated notebooks, and compete with other data scientists. In this rendition of ExploreML, we will get acquainted with Kaggle through a hands-on session of the ins and outs of the platform.
Oct 2021
Git.Set.Go will help you cover the topics of Git and we will help you find your very first contribution to open source with a PR. This will indeed give you a kickstart on how to apply your wonderful programming language skills on real time applications like Django and Flask or Node.js and React.
Jul 2021
In this podcast of ‘Let’s Talk ML’, students participating will be talking about their journey in the domain of Machine Learning. Wanting to support every listener in their journey, we will put out resources that help them get started. Apart from this, we will bring forward projects that we have worked on to give the listeners an idea as to what all they can build in this domain! The podcast will be wrapped with an insight of the capacity of this domain in the near future.
May 2021
TRIEPAD is an intelligent notepad built as a Data Structures & Algorithms project. It uses trie data structures to predict the next word and provide real-time autocompletion, enhancing typing speed and efficiency.
May 2021
A Discord bot built with Python and TerraForm to create cloud-based virtual machines on AWS, Digital Ocean, and others.
Dec 2020
This demo showcases our tool that helps identify the subtle, subconscious biases embedded in written articles. By providing an unbiased analysis, it empowers readers to better understand the author's perspective and make more informed interpretations.
Oct 2020
Using ROS, turtlebot simulation and state of art voice recognition and Natural Language Processing capabilities we have created and implemented a stack which can recognise the text from the speech and do a intent classification,to give commands to our robot. A small submap of the destination room is displayed and then the user can pinpoint a desired location in the room where robot is expected to reach.The robot then intelligently plans the path, with the help of advanced path planning navigates to the desired location.
