The nine questions
The same nine questions, answered twice
Numbers in brackets point to the numbered sources at the foot of the page. The line under each question says what a complete answer contains.
1 · Where must the data live for questions and agents to work?
A complete answer names the systems the data has to be copied into, and the systems that are read where they stand.
DatabricksGenie Agents work on tables, views and metric views registered in Databricks' Unity governance layer, with up to fifty objects allowed per agent though five or fewer is the recommended limit for accuracy. Lakehouse Federation adds a read-only route to outside warehouses such as Snowflake and BigQuery [1][2].
Lima JunoYour rows stay in the systems that hold them. Juno registers lakehouse tables beside SAP, Oracle and planning tables and reads each one in place, joining all of them in one answer. 2 · How far does it reach into SAP, Oracle and planning systems?
A complete answer names the interface used for each system and says which of them are live today.
DatabricksLakehouse Federation's read-only source list names Oracle, SQL Server, Snowflake, BigQuery and Teradata; SAP is missing from that list on the pages reviewed. MCP connectors, generally available from September 2026, and a Copilot Studio integration give Genie Agents a further route to outside systems [2][3].
Lima JunoMore than 40 connectors cover SAP ECC and S/4HANA, Oracle Fusion and E-Business Suite, planning suites, warehouse systems, lakes, databases, files and REST services. One question can join several of them, and the one agent in production, the PO cancellation agent, runs on federated ERP data the same way. 3 · How is the shared meaning, the semantic model or ontology, built, and by whom?
A complete answer says what arrives ready, what your team builds, and how long that modelling work takes.
DatabricksA Genie Agent is hand-curated by a SQL-proficient analyst, within limits of 100 instructions and 200 knowledge snippets per agent; Genie Ontology then infers a knowledge graph of business terms, entities and measures automatically from that curated set. Databricks' Unity metric definitions reached general availability at the 2026 Data and AI Summit [4][5].
Lima JunoLima’s supply chain ontology arrives filled in: materials, plants, suppliers, orders and stock. Juno proposes a mapping for each registered column with a confidence and a reason; your team confirms, adjusts and extends it. 4 · How do business users ask questions, and is the working shown?
A complete answer shows the question, the query that ran, the sources read and the access rules applied.
DatabricksBusiness users ask Genie One a question in plain language; agent mode, generally available from July 2026, plans multi-step research and cites the definitions it applied. The generated query is shown beside the result, and an Inspect feature checks the filters, dates and joins it used [1][6].
Lima JunoPeople ask ChatLima in plain language. Each answer shows its working: the ontology terms used, the systems read, the seconds taken and the query itself, under the asking person’s row and column access. 5 · How are agents and rules defined, and by whom?
A complete answer says who writes a rule, in which language, and how it is reviewed before it runs.
DatabricksAn analyst curates each Genie Agent from instructions, example query pairs, trusted assets, synonyms and join definitions, checking value matches across up to 100 million rows; thumbs-up and thumbs-down feedback then refines its answers over time [4][7].
Lima JunoA planner writes the policy in plain language, and the Planner Agent turns it into a reviewable planning table of tasks, data and rules. The PO cancellation agent, built this way, runs in production; Lima’s agent blueprint designs 960 further roles across five chains, the other supply chain task agents among them, yet to be built. 6 · How is autonomy set, and how do approvals work?
A complete answer gives the levels available, who approves what at each one, and how a level is raised and lowered.
DatabricksAccess runs through each person's own identity in Databricks' Unity governance layer, with row filters and column masks enforced per user and the agent author's compute credentials embedded in every run; a graded, numbered autonomy setting for actions stays unstated on the pages reviewed [3][8].
Lima JunoEach agent carries a level you set, A0 to A4. At A2 a named person approves every action; at A3 the agent acts inside value, budget and confidence limits and sends exceptions to people. The level is one of eight settings on an agent, each set by a named person. 7 · How do actions reach the ERP, and is each write confirmed?
A complete answer names the interface for each ERP and says where the acknowledgement of a write is kept.
DatabricksScheduled tasks reached general availability in June 2026, functions inside Databricks' Unity governance layer became callable tools in August, and MCP connectors followed in September; whether a write to an outside ERP carries a confirmed acknowledgement stays unstated on the pages reviewed [3].
Lima JunoWrite-back uses each ERP’s own supported interfaces: released APIs and BAPIs for SAP, REST resources and interface tables for Oracle. It starts per application after a confirmed test write, and every write keeps the acknowledgement. The PO cancellation agent in production writes back this way, under human approval. 8 · How does it learn from people’s decisions, and what is audited?
A complete answer says what is recorded for each decision and how that record shapes the next recommendation.
DatabricksFeedback such as instructions, example queries and thumbs-up or thumbs-down marks refines each Genie Agent over time, and Databricks reports that customers using its governance tools put twelve times more AI projects into production. A record tying one person's individual decision to a later audited outcome stays unstated on the pages reviewed [4][9].
Lima JunoEvery accept, edit and override is stored with the rule version, the facts at that moment and the outcome, so the next recommendation improves. The audit trail holds signal, reasoning, limits, approval, action and acknowledgement for each run; in production, 100% of the cancellation agent’s actions trace to a rule. 9 · What is the pricing unit, is there a trial, and where do analysts place it?
A complete answer names the unit of charge, the route to try it, and the analyst reports that cover the category.
DatabricksGenie One bills by Databricks Unit consumption, with 150 free units included per user each month worth about $10.50 in the US East region, and compute billed apart from that; a 14-day trial is offered. IDC names Databricks a Leader in its 2025-2026 MarketScape for Unified AI Governance Platforms [10][11][12].
Lima JunoJuno is licensed with the Lima platform and priced per engagement. The token manager budgets and meters every model call, per agent and per team. Twenty use cases are worked through on this site, one page each, and the analyst guide sets out the markets these products sit in. 