Databricks Introduces Genie Code for AI Runtime in Public Preview
Databricks has launched Genie Code for AI Runtime into Public Preview, offering new capabilities for developing and troubleshooting deep learning workloads within notebooks. This tool automatically generates distributed training code, resolves common environment and dependency issues, and assists in debugging GPU and distributed workload failures. It is designed to streamline the deep learning development process for data scientists and ML engineers using Databricks AI Runtime.
Features (1) ›
- Genie Code for AI Runtime enters Public Preview
Genie Code assists developers and data scientists in creating and debugging deep learning workloads within notebooks connected to AI Runtime. It can generate distributed training code, diagnose and resolve environment and dependency problems, and help debug issues related to GPU and distributed workloads.
https://docs.databricks.com/aws/en/release-notes/product/2026/august#genie-code-for-ai-runtime-public-preview
Related releases
- Lakebase default storage quota increased to 32 TB Databricks Release Notes ·
- Databricks Lakeflow Connect: Zerobus Ingest to Streaming Tables GA Databricks Release Notes ·
- Databricks Genie One Introduces Desktop App, Enhanced Collaboration, and Expanded Context Databricks Blog ·
- Databricks Lakeflow Designer August Release Adds Unique Operator and Canvas Enhancements Databricks Release Notes ·
- Databricks Introduces Unified Trace Table for AI Gateway Monitoring (Beta) Databricks Release Notes ·
- Indra Unifies EV Charging Data on Databricks, Enabling Self-Service Analytics Databricks Blog ·