---
title: "What’s New in the AI Platform: Agents for ML Engineering, Our Deep Learning Platform, and New Capabilities for Real-Time ML"
description: "The announcement presents three additions to the Databricks AI Platform: Genie Code support for ML engineering, AI Runtime’s serverless GPU environment, and expanded real-time ML capabilities. Genie Code integrates with Unity Catalog, Feature Store, training, serving, monitoring, and MLflow, assisting with feature engineering, model training, deployment, evaluation, and production operations. AI Runtime provides on-demand serverless NVIDIA A10 and H100 GPUs, supports high-performance multinode training with RDMA and high-performance data loading, and adds Lakeflow Jobs, DABs, MLflow, and Unity Catalog integration. For real-time ML, the platform adds declarative feature engineering, streaming features, online feature serving on Lakebase, and enhanced Model Serving targeting 300K+ QPS with under 10ms p99 latency overhead. Reported customer examples include faster workflows, lower infrastructure costs, and production scaling beyond 100K QPS."
---

# What’s New in the AI Platform: Agents for ML Engineering, Our Deep Learning Platform, and New Capabilities for Real-Time ML

[Databricks](https://yomu.fyi/company/databricks) · Tejas Sundaresan, Mike Del Balso · Jun 17, 2026

**Type:** Announcement

## Summary

The announcement presents three additions to the Databricks AI Platform: Genie Code support for ML engineering, AI Runtime’s serverless GPU environment, and expanded real-time ML capabilities. Genie Code integrates with Unity Catalog, Feature Store, training, serving, monitoring, and MLflow, assisting with feature engineering, model training, deployment, evaluation, and production operations. AI Runtime provides on-demand serverless NVIDIA A10 and H100 GPUs, supports high-performance multinode training with RDMA and high-performance data loading, and adds Lakeflow Jobs, DABs, MLflow, and Unity Catalog integration. For real-time ML, the platform adds declarative feature engineering, streaming features, online feature serving on Lakebase, and enhanced Model Serving targeting 300K+ QPS with under 10ms p99 latency overhead. Reported customer examples include faster workflows, lower infrastructure costs, and production scaling beyond 100K QPS.

## Context

The announcement addresses the time and operational complexity involved in moving ML systems from experimentation to production, including repetitive lifecycle work, GPU infrastructure management, and the difficulty of maintaining low latency and high throughput as real-time traffic changes.

## Approach / What changed

It introduces Genie Code integrations across the Databricks ML lifecycle, AI Runtime as a serverless GPU and multinode training environment, and Feature Store and Model Serving enhancements for real-time workloads. The capabilities include Unity Catalog and MLflow integration, Lakeflow orchestration, RDMA and optimized data loading, streaming and declarative features, Lakebase online feature serving, and automated serving adaptation.

## Takeaways

- Genie Code uses context from Unity Catalog, the Databricks ML stack, and MLflow workflows to support feature engineering, training, deployment, monitoring, drift detection, and production operations.
- AI Runtime offers serverless NVIDIA A10 and H100 GPUs with no cluster management, pay-per-use billing, high-performance multinode training, and Lakeflow orchestration for long-running GPU workloads.
- Enhanced Model Serving targets 300K+ QPS with under 10ms p99 latency overhead, while the announcement reports customers scaling beyond 100K QPS and cutting infrastructure costs by up to 90%+ versus self-managed stacks.

**Tags:** [AI Agents](https://yomu.fyi/topic/ai-agents), [GPU](https://yomu.fyi/topic/gpu), [Lakebase](https://yomu.fyi/topic/lakebase), [Model Serving](https://yomu.fyi/topic/model-serving)

- Source: [Databricks](https://www.databricks.com/blog/whats-new-ai-platform-agents-ml-engineering-our-deep-learning-platform-and-new-capabilities)
- Source URL: https://www.databricks.com/blog/whats-new-ai-platform-agents-ml-engineering-our-deep-learning-platform-and-new-capabilities
- Ingested by Yomu: 2026-08-30T17:01:37.914Z

[Read original post](https://www.databricks.com/blog/whats-new-ai-platform-agents-ml-engineering-our-deep-learning-platform-and-new-capabilities)