• Traced critical request path among 18 microservices in a distributed system, created Grafana Dashboards for 50+ gRPC method performance, and drove a company-wide initiative for CUJ-based SLO adaptation

• Developed availability and latency SLOs using in-house SLO tooling for the Reddit site, adding metrics, Grafana dashboards, and alerting rules based on Google SRE best practices to help engineering teams for better incident response

• Identified key SLIs, and SLOs and instrumented Prometheus into Reddit’s service catalog to enhance its observability, then analyzed performance and increased its data freshness by 33%

• Analyzed Kubernetes instance lifecycles to uncover the fundamental origins of GraphQL 5xx error

anomalies within a distributed system

• Implemented functionalities to create new robot caregiving workflows and comment on existing ones

through the web interface using the MERN stack

• Managed software development life cycle for Reddit's in-house service catalog, incorporating Spotify's open-source service Backstage and consolidating data from GitHub, Slack, Sentry, and other sources into a unified, trusted system used by 500+ engineers

• Enrich the service directory with metadata stored in AWS S3 with a Python script

• Created reusable React components to display metadata in frontend in TypeScript,

increasing micro-services observability and accountability