Architecting scalable AI infrastructure and high-throughput data platforms to automate system diagnostics and enhance event discovery across distributed software services using vector-based search and agentic workflows.

* Developed version control for Drag & Drop UI and API editor, enabling users to

efficiently track changes, manage revisions, and collaborate.

• Implemented a Role-Based Access Control (RBAC) system, enabling granular

resource access and enhancing security and user management flexibility.

I worked as a Backend Engineer on an AI-driven Enterprise Distributed Document Processing Platform. Over the course of six years, I contributed to multiple core components of the platform, playing a key role in the company's growth as it scaled from around 50 to over 250 employees. Sample projects:

* Designed and implemented on-site ML model training, enabling customers to efficiently handle unstructured data and extract valuable insights, improving their data processing workflows.

* Developed and optimized distributed data processing pipelines, enhancing system scalability and performance. Improved public API functionality, leading to faster data retrieval for customers, and created a near real-time dashboard that streamlined decision-making processes.

* Implemented Human-In-the-Loop (HITL) flow and data collection, improving data labeling efficiency and ML model accuracy for more reliable outcomes.

I have published several peer-reviewed research papers on detecting solar activity in 4K satellite images using computer vision. Additionally, I developed a Postgres/PostGIS plugin for efficiently representing and analyzing spatio-temporal data derived from solar activity.

I was part of the IoT team, where I primarily worked on an Android application that communicates with smart appliances. In addition to contributing to the embedded application, I also developed a Java-based REST API for monitoring the status of these smart devices.