The run-sql endpoint now supports a codeType parameter to distinguish between SQL and Python code execution. This enables M2M clients to execute Python code while maintaining SQL validation for regular queries.
The account API response now includes a reference to the account's observability workspace, enabling clients to detect when observability features are available.
Create and manage reusable AI skills that can be triggered via slash commands in chat. Skills allow you to define custom prompt templates that users can invoke to perform specialized tasks, such as SQL query generation, data analysis, or domain-specific assistance.
Data assets can now have a postgres_table type in the externalMeta.type field, enabling PostgreSQL tables to be properly identified and managed alongside existing Snowflake and Sledhouse assets.
Connect directly to PostgreSQL databases for AI-powered analytics. Bobsled AI now supports PostgreSQL alongside Snowflake and Sledhouse, enabling natural language queries against your existing PostgreSQL infrastructure.
Connect your Snowflake account directly to Bobsled AI using secure OIDC authentication, bypassing Sledhouse infrastructure entirely. This gives you full control over data residency while enabling AI-powered analytics on your existing Snowflake investment—no data copying or ETL required.
Ask data questions directly in Slack by @mentioning Bobsled AI in any channel. Teams can now get answers without context-switching to a separate tool, making data insights accessible in the flow of work where decisions happen.
Enable or disable the AI's ability to supplement data answers with web search results on a per-workspace basis. When enabled, the AI can look up current information—like exchange rates, industry benchmarks, or company news—to provide richer context alongside your data insights.
Automatic metadata collection enriches the AI's understanding of your data assets with column statistics, row counts, and sample values. This context helps the AI generate more accurate SQL by understanding data types, cardinality, and common values—reducing query errors and improving natural language interpretation.
Query across multiple data domains in a single conversation by creating composite workspaces that combine semantic models from different source workspaces. This enables cross-functional analytics—like joining sales data with marketing campaigns—without duplicating data or maintaining separate models.
