kkkgabriel

- Lead a team of 3 to design and build a new service that automates the handling of common oncalls and alerts - Head the deployment of model distribution and serving stack in multiple new data centers - Deployed new features on more than 100 services with internal AB testing framework based on model performance - Managed GPU resource allocation to clients in Asia Pacific, Europe and US regions - Assisted in automating GPU resource quality control process to reduce online issues - Resolved oncalls, alerts and business requests for model distribution and serving cases

- Supported operations in cryptocurrency space such as bot trading, order management, deal settlements, accounting, and fund transfers - Monitored applications with tools such as Grafana, Loki, Prometheus and shell scripting for 24/7 uptime - Worked with CI/CD tools such as Kubernetes, teamcity, chartmuseum, rundeck to maintain deployments - Developed services to account for daily P&L and outstanding settlements