Aditya Karad

I am a Member of Technical Staff at Oracle Database, where I work on real-time replication, CDC, and other database internals. I've previously worked in ad-tech at AppLovin and fin-tech at JP Morgan.

Oracle Database work

I work on two products at Oracle Database:

  1. Oracle GoldenGate is a real-time CDC tool. My work spans Replicat, which applies captured database changes while preserving transactional correctness; Credential Store; and Admin Client, the CLI used to manage a GoldenGate instance.
  2. In Oracle Database, I work on XStream APIs for developers building CDC integrations, Apply Server, which serves a similar role to Replicat, and ACDR APIs for customizable automatic conflict detection and resolution.
  • I redesigned GoldenGate's Credential Store to support Kerberos, OAuth, passwordless, and plugin credentials for database connections, replacing a legacy design that did not support them. I now solely maintain this component. The design preserves backward and forward compatibility, supports future authentication methods, and accounts for security issues coding agents can uncover.
  • I implemented batched execution for specific DML operations in Oracle Database's XStream-In Apply layer, improving replication performance on TPC-C workloads.

Background

I completed my master's at UC Irvine with a focus on systems, graduating early with a 4.0 GPA. My coursework included database systems, distributed systems, operating systems, data structures, and cryptography. I earned my bachelor's in Electrical Engineering at the Indian Institute of Technology Roorkee, where I studied microprocessors and hardware. At UC Irvine, I worked with Professor Chen Li on Apache Texera and helped operate isolated Texera deployments with Kubernetes and AWS infrastructure.

Before Oracle, I worked as a backend software engineer at AppLovin, where I built an asynchronous event pipeline using Google Pub/Sub and BigQuery and worked on performance analysis in a multithreaded Java service. Earlier, I was a software engineer at JPMorgan Chase, where I worked on cloud migrations and regulatory reporting software. I have also contributed patches to Apache AsterixDB.

Projects and interests

My projects include a C++ database implementation based on CMU's BusTub coursework, a fault-tolerant restaurant ordering system using RabbitMQ and MongoDB, a Rust shell-history tool, a Kubernetes control plane for Apache Texera, and several small products such as EarlyApply and Kindle Export to Notion.

  • I'm interested in distributed systems, databases, and algorithms.
  • You can see my past projects on the projects page and what I'm up to nowadays on the updates page.
  • I like reading about AI, systems, and coding agents in engineering blogs. I keep links and notes in my reading log.

Machine clients can request this URL with Accept: text/markdown or start with llms.txt.