Databricks

Integrated training observability streams logs and simplifies distributed model debugging.

Connects generative protein design with governed data and serverless GPUs.

Databricks supports experimentation, customization, and serving of Nemotron models.

Validated workflow customizes time-series anomaly detection models using enterprise data.

NVIDIA NeMo integration enables open model customization on serverless GPUs.

Custom container environments support GPU model training with NVIDIA software.

AI agents unify customer records through normalization and identity resolution.

AI agents build audiences and personalize campaigns using governed customer data.

Persistent memory for AI agents using managed sessions and retrieval.

Extracts specified document fields with optional citations and confidence scores.

SQL-native classification predicts categorical values using managed machine learning models.

SQL-native regression predicts continuous values using centrally maintained machine learning.

SQL-native machine learning detects unusual observations in temporal datasets.

Combines vector similarity and keyword search inside application Postgres databases.

Integrates Meta's ads MCP server with Databricks for AI-driven marketing.

Enables personalized, permission-aware AI experiences in Databricks Apps.

New SQL join for batch vector search in Databricks.

AI-powered tool for analyzing data and forecasting trends.

AI-powered candidate retrieval for personalized product recommendations.

Always-on engagement loops for continuous marketing optimization.

Backend agents for automating fund operations and NAV reconciliation.

AI coworker for asset management finance with governed ontology.

Fast decision-making AI function for governed data.

Unified platform for building and optimizing agent fleets.

AI assistant for interactive dashboard queries and insights.