I have played a pivotal role in developing AI-driven solutions that enhance product configuration, streamline workflows, and optimize client engagement. I designed and implemented a full-scale ETL and training pipeline using FastAPI, PyTorch, Vertex AI, and PostgreSQL, which significantly reduced training time by 80% and increased client engagement by 50%. One of my key achievements was developing a Neural Network based recommendation system, which enabled accurate product configurations and directly contributed to onboarding major clients. I also explored and implemented cutting-edge research in Tab Transformers and VQ-VAE, improving product configuration accuracy over the baseline model. To further enhance system efficiency, I automated model monitoring and undeployment workflows using Google Cloud Composer (Apache Airflow), reducing operational costs by 70%. Additionally, I built a voice-activated transaction editor powered by LLMs, cutting down editing time for large transactions by 50%, and developed a Retrieval-Augmented Generation (RAG) system with FastAPI and LlamaIndex, improving internal documentation search and query efficiency.

During my time at Urban Company, I was a member of the matchmaking team, which was tasked with identifying the most suitable partner for customers who submitted requests through the application. While working there, I had the opportunity to develop my expertise in various technologies, such as SpringBoot, Java, NodeJs, Typescript, Machine Learning, PySpark, Python GoogLeNet, MongoDB, AWS, Redis, Kafka, RestFul Web services, Docker, and Kubernetes. I worked on challenging assignments that demanded scalability and reliability. I have a documented record of successfully completing intricate projects and to a high standard and have been acknowledged for my dedication, accountability, and ownership.

I had a wonderful experience working at this place. The culture at UC is exceptional and has not only contributed to my professional development but also helped me to develop as a person. The company instills important values such as taking ownership, being proactive, daring to take risks, simplifying processes, prioritizing customer satisfaction, and fostering teamwork.

During my tenure at Samsung Research, I was a member of the SystemUI team. My responsibilities involved porting Android from an earlier version to a newer version and resolving issues with Samsung devices. I have resolved 500+ bugs during my tenure at Samsung. At Samsung, I gained skills in Spring, Java, Machine learning. I filed an A2-grade patent after developing a tensorflow lite-based XGBoost model for detecting Grip on mobile devices, significantly reducing the requirement for Grip sensor hardware in Samsung mobile devices. I also developed and deployed a Random Forest classifier Machine learning model during a hackathon to efficiently route bug tickets to the right team, reducing manual tester effort and erroneous bug assignments.

Using Spring, I created a backend to fetch and process InMobi stories, and store them in database. These stories were then retrieved and displayed on the Samsung lock screen, resulting in increased user engagement.