Create

Sign in to ReadmeX

or

How we handle your data: Privacy Policy

ReadmeX

Your AI reading companion

What would you like to catch up on?

Pick a briefing, or ask your own question.

Your community and people briefings stay personal to you.

Unity Gateway

AI overview

Sign in and the AI will write an overview from our coverage.

Headlines · 5

  1. SAP and Databricks expand AI capabilities in SAP Business Data Cloud

    SAP and Databricks announced an expanded partnership that plans to bring Genie, Lakeflow, Unity Gateway and Lakebase to SAP Databricks within SAP Business Data Cloud. The companies also plan a bidirectional Genie-Joule integration, while SAP Application Data Foundation makes SAP application data available without custom extraction pipelines. SAP Databricks and SAP Business Data Cloud Connect for Databricks are available, but no release dates were provided for the announced additions or the planned integration.

    Databricks Blog · 🔥 9
  2. Why Wealth Management Onboarding Needs More Than AI

    Databricks argues that accelerating document extraction, screening or routing does not resolve the handoffs and fragmented case information that slow wealth management client onboarding. It proposes a governed architecture in which supervisor agents coordinate specialist agents, approved tools and human reviewers, with Unity Catalog and Unity Gateway providing access controls, audit trails and runtime guardrails. The article cites industry studies to illustrate the scale of onboarding delays and operational fragmentation, but its platform capabilities are presented from the vendor's perspective.

    Databricks Blog · 🔥 4
  3. Databricks engineer: AI cost management shifts from tokenmaxxing to valuemaxxing

    Databricks published an interview with Kevin Stumpf, a senior staff engineer and technical lead for its Unity Gateway, arguing that enterprise AI observability is moving from measuring token volume to measuring return on investment. Stumpf describes AI cost management as measuring, attributing, controlling and optimizing spend across models, agents, applications and teams, and says the industry is moving from "tokenmaxxing" to "valuemaxxing." He recommends hard spend caps set above normal power-user levels, differentiated budgets by team or project, alerts before caps, and shifting requests to open source models once a team has used about 70% of its budget.

    Databricks Blog · 🔥 0
  4. Databricks launches Beta Agent Memory for Agent Bricks agents

    Databricks introduced Agent Memory on Agent Bricks, its agent developer platform; it is available in Beta and runs on Lakebase, Databricks' serverless Postgres. It is built on the MemCurrent architecture from Databricks Research, combining plugins that log and load context automatically, tools the agent calls on demand, and an offline "dreaming" pipeline that consolidates memory in the background. Databricks says its internal benchmarks tripled agent pass rate on real enterprise tasks while using 31% fewer tokens, vendor-run figures that have not been independently verified.

    Databricks Blog · 🔥 0
  5. TinyFish runs its web-native agent model on Databricks AI Runtime

    TinyFish is developing a specialized web-native model for enterprise agents that work across the live web, and has consolidated its post-training, evaluation and production serving on Databricks: managed GPUs via AI Runtime, evaluation with MLflow, and deployment through Model Serving. The team started with supervised fine-tuning and is moving toward Direct Preference Optimization. On its internal go/no-go benchmark, the model completed 1,691 of 2,178 web-agent tasks for a 77.6% task-success rate, with a separate LLM judge reviewing the request, action history and final result; partial completion or incorrect answers count as failures. Over a recent 15-day period, TinyFish recorded 294,781 agent runs involving the model across 6.4 million tool steps, at an estimated serving cost of $0.088 per run and $0.0040 per tool step — covering only that model, not the full agent stack. TinyFish already serves its search reranker and query rewriter on Databricks and uses Unity Gateway to centralize access to external foundation models.

    Databricks Blog · 🔥 0

Experience and discussion from the community

Share my Unity Gateway experienceAsk about Unity Gateway

Nobody has shared their experience with Unity Gateway yet.