Story
Snowflake
Goldman SachsHow Goldman Sachs Runs Applied Machine Learning on Snowpark
Goldman Sachs, a financials organization in the United States, uses Snowpark from Snowflake to support applied machine learning for data scientists and ML engineers.
When feature pipelines outgrew homegrown tracking, Goldman Sachs brought in Snowflake.
“Goldman Sachs is based in the United States and runs financials operations at a scale where feature pipelines cannot live in side channels.”
“Data scientists and ML engineers were reconciling competing copies of the same work, which slowed applied machine learning and hid who owned the next step.”
Independent write-up. Figures come from public sources or are illustrative.
Value results
| Category | Value result |
|---|---|
| Capability | Named workflow replaces ad hoc routing for feature pipelines |
| Capability | Applied machine learning stays visible to adjacent teams through Snowpark |
| Capability | Data scientists and ML engineers work from the same Snowpark record for feature pipelines |
Story
Goldman Sachs is based in the United States and runs financials operations at a scale where feature pipelines cannot live in side channels. Data scientists and ML engineers were reconciling competing copies of the same work, which slowed applied machine learning and hid who owned the next step.
The company runs applied machine learning on Snowflake, with Snowpark as the product data scientists and ML engineers actually open. Snowflake is a cloud data platform that separates storage and compute so organizations can share, analyze, and govern data across clouds. For Goldman Sachs, that means data scientists and ML engineers can open one workflow, see feature pipelines, and let neighboring teams join without inventing a parallel stack.
Public materials confirm the companies and products. They do not always publish a single verified KPI for this pairing, so the outcome here is operational: clearer ownership, fewer stalled handoffs, and a shared record for feature pipelines.
