Two microservices developed in Spring Boot. Implemented CI/CD pipeline in GCP using Cloud Build, Cloud Source Repository and Artifact Registry. Created Kubernetes cluster using terraform.

Created cloudbuild.yaml which will package the source code, containerize it and deploy the workload to Kubernetes cluster using deployment yaml files, when a new commit is pushed to the repository.

Attached persistent volume to the kubernetes cluster accessible by both microservice containers. Exposed one microservice as a service to the internet.

Used convolutional neural networks (CNN) to identify images as 'dogs' or 'not dogs'.

Evaluated the algorithm's effectiveness in correctly identifying a dog's breed.

Timed the algorithms to balance accuracy with runtime efficiency.

Loaded a pre-trained GPT-2 model and evaluated its performance.

Performed parameter-efficient fine-tuning using the PEFT library with LoRA configuration.

Conducted inference using the fine-tuned model and compared its performance to the original model.

Built a web platform aimed at improving team collaboration and productivity through effective task management. Features include:

User Authentication: Secure login system using AWS Cognito.

Data Storage: Persistent data storage and retrieval using SQL databases. with AWS RDS

Deployment: Leveraged AWS services for robust deployment, including Elastic Beanstalk for application hosting, EC2 for scalable computing power, Load Balancers for traffic management, VPC for network isolation, and RDS for relational database services.

Used AWS Cloudformation to deploy AWS resources

Contributed to the development of Eduhub, focusing on the online testing feature which includes:

Question Banks: A repository of questions that can be categorized and reused in various tests. User is able to do CRUD operations on Questions

Test Creation: Allows educators to create customized tests with varying levels of difficulty and formats. User is able to do CRUD on tests.

Test Attempts: Enables students to take tests within a controlled environment and secure environment

Result Generation: Automated evaluation and result generation providing immediate feedback to students

Developed the backend of a social media and events platform with robust features to enhance user interaction and engagement:

Authentication: Secure login and registration system to protect user data by using Spring Security and JWT

User Profiles: Comprehensive profile management allowing users to customize their online presence.

Posts (CRUD Operations): Full suite of Create, Read, Update, and Delete functionalities for user posts.

Chat Functionality: Real-time messaging system to facilitate communication between users.

Worked in Test Driven Development and Agile Environment with Code Coverage of 85%

This mobile application was created to streamline the process of renting and buying pre-owned items. I worked on

Calling System: Allows users to directly contact sellers or renters via in-app calling using Zegocloud

User Authentication: Ensures secure login and registration processes using firebase AUth

User Profiles: Enables users to create and manage personal profiles with their listings and preferences.

Homepage: Features a user-friendly interface displaying the latest items available for rent or purchase.

We developed a mobile application designed to enhance the travel experience for tourists. This comprehensive tool offers several features to ensure a seamless and enjoyable journey

Recommendation System: Provides personalized suggestions for places to visit, eat, and stay based on user preferences and location.

Tourist Support Chatbot: An AI-powered chatbot available 24/7 to answer queries, offer travel tips, and provide real-time assistance.

Translator: Facilitates easy communication with locals by translating text and speech between multiple languages.

Storytelling Feature: Delivers rich historical and cultural information about various attractions through engaging narratives.