Consuming Legend Data Models & Services Using BI, Python/ML & othe...
summary
Goldman Sachs open-sources a usable pipeline: a JSON TDS protocol that models tabular queries, a Python Pure library and a SQL connector that integrate with Pandas, Trino and BI tools, and Zipkin tracing for observability. Demos run against a 1 million-row vehicle complaints service, push filters/sorts/aggregations down to the database, return small result sets, and show materialized SQL — functionality works and Q4 release targets protocol, execution and client libs.
AI-written summary · about
chapters (9)
- 0:00Start
- 3:32Data Modeling
- 4:24Problem Space
- 6:03Data Model
- 9:36Client Libraries
- 12:59Code Samples
- 14:23Sql Operations
- 16:44Demos
- 17:24Jupyter Notebook
related talks
36mAn Open-sourced Solution to Data Governance? How Legend May Be t...
18mElevating enterprise data through open source LLMs
22mReliable data engineering made easy
40mPyData Stack: Pure Python open source data platforms
01h 26mData Driven Action: a primer on Data Science
09mSponsored Keynote: Building a Financial Lakehouse with Legend
43mData warehousing reinvented for today's needs
24mData Pipelines to Dashboards