Specializing in SQL performance tuning
As a support technical lead at Snowflake Melbourne, Eric Lin has over 10 years of troubleshooting and tuning SQL performance issues, covered around 5+ year for Hive, Impala in Cloudera, to also around 5+ years on the Snowflake platform. Eric also oversees the APAC support operations from the technical perspective, helps developing tools and processes to better support our support engineers on their day to day tasks, as well as working closely with global tech lead team and higher support, engineering and product management team to improve various functions within the whole organization.
Every engineering team eventually hits the same wall: data scaling issues that slow down applications and cause cloud bills to skyrocket. CoCo is Snowflake's AI-powered assistant but it does way more than just autocomplete your code. Because it has direct access to Snowflake's metadata and query profiling tools, you can turn it into an interactive troubleshooting partner that lives right inside your IDE. The real secret weapon? CoCo Skills. These are modular, domain-specific instructions that teach the AI how to hunt down specific performance killers. In this talk, we'll build purpose-built CoCo skills live on stage. You'll see exactly how to train the assistant to diagnose: Query spilling and partition pruning Join order regressions and data skew Warehouse queuing and resource bottlenecks Whether you're a data engineer tired of getting "Why is my query slow?" Slacks, or a platform lead trying to scale your infrastructure, you'll walk away with a practical framework. Learn how to encode your team's tribal knowledge into reusable, shareable CoCo skills that turn painful root-cause investigations into guided, repeatable wins.
