Jan 2025 – Present · 1 yr 4 mos · Bengaluru, Karnataka, India

Sep 2023 – Dec 2024 · 1 yr 3 mos · Bengaluru, Karnataka, India

• ● Reduced Costs and Optimized Performance: Achieved significant resource cost savings and system optimization by improving Spark job performances across multiple domains. Reduced Bootstrap Job resource usage by 80% and Batch Job resource usage by an extraordinary 4200%. Also contributed to the activity processing redesign, including designing the Backfill Event Scheduler as an interim solution to resolve lead activity ordering issues.
• ● Enhanced Observability for DC Services: Spearheaded initiatives to improve service reliability and monitoring, including fine-tuning alert thresholds, creating automated CSOs, and adding pipeline lag alerts. Developed detailed dashboards for enhanced visibility and integrated custom notifications into core services to inform customers and developers of system failures. Built a reusable notification framework for seamless integration with other services, improved OnCall Reference Guides, and strengthened operational efficiency and customer satisfaction.

Jan 2023 – Jun 2023 · 5 mos · Noida, Uttar Pradesh, India · Hybrid

• ● Developed a Customer-Facing Plugin for Adobe Express Using LLMs: Created a plugin for Adobe Express that leveraged LLMs to enable users to seamlessly create presentations from either a file or a provided topic. The generated slides had dynamic texts as well as designs, based on the content required by the user.
• ● Fine-Tuned Large Language Models for Long-Form Question Answering: Optimized self hosted LLMs such as Flan-T5 and Bloom for the specific downstream task of long-form question answering over internal software documentation. The process involved curating datasets, designing training pipelines, and ensuring the models could generate accurate and contextually relevant answers. The whole Retrieval Augmented Generation implementation was done from scratch, including custome encoders and a personalised recommendation engine.

May 2022 – Jul 2022 · 2 mos · Bengaluru, Karnataka, India

• ● Built a tool that automated the process of trait creation for Adobe Audience Manager, reducing the time taken
• for the process to complete from hours to minutes, effectively increasing efficiency and productivity.
• ● Built an API based alert microservice for Adobe Audience Manager that automatically sent emails every time the platform experienced lesser usage than average, effectively reducing loss in business and quick root cause analysis.
• ● Led a team of other fellow interns in the development of the Intern Portal showcasing the entire journey of the
• team's internship, with coding, testing, analytics, and deployment done in a very strict time frame.