in Projects / Personal projects / Web-dev / Sockets
In this project, I created an online song guessing game using web sockets. The game is a multiplayer game where players can join a room and guess the song that is playing. The game is played in real-time and the first player to guess the song correctly wins the round. The game has a leaderboard that keeps track of the scores of all the players.
The Smart Fan is one of my IOT projects which used a temperature and humidity sensor to control the speed of a fan connected to it. It also incorporates a LCD screen to display the readings from the temperature and humidity sensor. The project was built using an Arduino Uno, a DHT11 sensor. Below is diagram of the project for the circuitry:
For this project I created a Simple Security Alarm using an Ultrasonic Sensor with a Buzzer and a LED screen to show the danger level. The diagram for the wiring is as shown below:
For the Microsoft AI Hackathon, I worked on a project called Polyglot Pals which is a language learning platform that uses the power of AI to help users learn a new language. The way it works is that it acts as a friend for the user, allowing the user to speak and write in the language of their choice and the AI will respond back in the same language. The AI is able to understand the context of the conversation and respond back in a way that is both helpful and informative.
For the Google AI Hackthon, we were tasked with creating a product that utilizes the new Generative AI model Gemini API. My team decided to create a task app which utilizes the Gemini API’s ability to understand and comprehend natural language to parse and generate tasks for the user. This allows users to instead of spending time thinking of what to call their tasks, filling in the details, and setting the due date, to just write a sentence or two about the task or copying and pasting notes about the task and let the AI do the rest.
The project was done as a further extension upon my Senior Capstone Project. Which I undertook the challenge due to my interest in the Deep Learning and hybrid models that were used in the project. In my Senior Project I worked on a Deep Kernel Learning model which combined the capabilities of a Gaussian Process with a Neural Network forming a more complex and improved model. However, there are many limitations to the model, such as it’s limited ability to incorporate the spatial and temporal dependencies of the climate data. This is where a Neural Process comes in, which is a model instead takes and blends ideas from neural networks and Gaussian processes.
Datahacks is an annual event hosted by the Data Science Society at UCSD, is a hackathon where students of all backgrounds come together to work on a project for 24 hours. This year the event was held on the 6th-7th of April 2024 at UC San Diego and was held in conjunction with ACM UCSD’s Diamond Hacks.