Designed and implemented environment variable propagation across different resources as a feature for Op.

Optimized resource count query with tag filters by transforming abstract syntax tree (AST) and pushing down the code logic to SQL level, achieving 2X to 4X speedup on a scale of 2000 resources.

Improved backend planner time from 10s to 0.1s in a large scale in‐memory cache miss scenario, by minimizing database calls, constructing local cache, and eliminating unnecessary data copy.

Conducted snapshot testing and automated the test cases for robust validation.

Iterated on an Image Analysis RPC Framework in both C++ and Python to support more types of operations. Migrated existing application into this framework with declarative pipelining to increase reusability.

Developed a data analysis tool in Go to retrieve log data of remote service calls from Kafka message queue, and generate daily reports, which significantly improved service operational visibility for the whole team.