KOBAI VS. AtScale

Both AtScale and Kobai sit between your data and your users. The fundamental difference: AtScale optimizes BI query performance and metric consistency. Kobai models enterprise knowledge and relationships to power AI and cross-domain intelligence.

 

Two Different Jobs for the Semantic Layer

 

The term "semantic layer" has come to describe two distinct but complementary capabilities:

 

 

AtScale structures data for reporting. Kobai structures meaning for understanding.

 

Purpose: BI Performance vs Enterprise Understanding

 

AtScale: Semantic Layer for BI Optimization

 

AtScale addresses the challenge of making BI tools faster and more consistent:

• Query acceleration: Aggregate awareness and intelligent caching to speed dashboard performance

• Metric definitions: Centralized business metrics so teams stop debating KPI calculations

• BI tool compatibility: Connects to Tableau, Power BI, Looker with optimized SQL generation

• Data virtualization: Query across data sources without moving data for BI consumption

This solves a critical analytics problem: consistent, performant BI at scale. AtScale is purpose-built for the analytics use case.

 

Kobai: Semantic Layer for Knowledge and Context

 

Kobai addresses a different challenge: making enterprise data intelligible by modeling what things mean and how they relate:

• Ontology-driven modeling: Define real-world entities (assets, parts, suppliers, events) and their relationships

• Cross-domain intelligence: Connect design → manufacturing → maintenance → operations across system boundaries

• Relationship-aware queries: Navigate multi-hop questions without writing complex joins

• AI grounding: Provide structured context and relationships that make AI answers traceable and trustworthy

This solves a knowledge problem: helping organizations understand what their data means in business terms and how entities connect across domains. Kobai is purpose-built for the understanding use case.

 

Why This Distinction Matters: Analytics semantics (AtScale) and knowledge semantics (Kobai) solve different problems. Most enterprises need both but they serve different moments. AtScale makes existing BI workflows faster and more consistent. Kobai makes complex, cross-domain questions answerable by modeling the business context that BI tools don't capture.

 

What Gets Modeled: Metrics vs Entities and Relationships

 

AtScale: Metrics, Dimensions, and Logical Models

 

AtScale's semantic layer focuses on the constructs that BI tools consume:

• Business metrics: Revenue, conversion rate, customer lifetime value are calculated consistently

• Dimensions and hierarchies: Time, geography, product categories for slice-and-dice analysis

• Aggregates and rollups: Pre-computed summaries for dashboard performance

• Logical data models: Star schemas and dimensional models optimized for analytics queries

 

This modeling approach serves analytics workflows extremely well—it's designed for the questions BI users ask.

 

Kobai: Ontologies, Entities, and Semantic Relationships

 

Kobai's semantic layer focuses on modeling the real world and how things connect:

• Domain ontologies: Aircraft, assemblies, parts, suppliers, maintenance events, failures—business concepts as first-class objects.

• Typed relationships: 'Part X is installed in Assembly Y,' 'Supplier Z manufactured Component Q'—relationships with business meaning.

• Cross-domain identity: Resolve that 'Asset 12345' in ERP is the same physical object as 'Tail ABC' in MES and 'Unit XYZ' in maintenance logs.

• Lifecycle traceability: Track entities through design → manufacturing → operations → disposal with lineage.

 

Why This Matters: BI dashboards need aggregated metrics. AI systems and cross-domain investigations need structured knowledge about what exists and how it connects. AtScale models for reporting. Kobai models for reasoning. These are complementary capabilities, not competing ones. Many organizations eventually need both.

 

Primary Users: Analysts vs Domain Experts and AI Systems

 

AtScale: Built for BI Analysts and Dashboard Users

 

AtScale is designed to serve the analytics organization:

• Business analysts creating dashboards and reports

• Data teams managing BI infrastructure and query performance

• Executive users consuming consistent KPIs across tools

• Analytics engineers building dimensional models

These users benefit from faster queries, consistent metrics, and simplified access to BI-ready data.

 

Kobai: Built for Domain Experts, AI, and Cross-Functional Teams

 

Kobai is designed for people who need to understand context and relationships:

• Operations engineers investigating failures across the asset lifecycle
• Quality teams tracing defects through supply chains and manufacturing processes
• Compliance officers requiring traceable lineage from part to certification
• AI systems (copilots, agents) needing structured business context to ground answers
• Data product owners building knowledge-driven applications

 

Why This Matters: If your primary need is making BI faster and metrics consistent, AtScale serves that user base well. If your primary need is helping domain experts understand cross-system relationships or grounding AI in business context, Kobai serves that user base. The tools optimize for different workflows.

 

AI Integration: Query Acceleration vs Context Grounding

 

Both platforms position themselves as enabling AI workflows, but they approach the problem differently.

 

AtScale: Accelerating AI-Generated BI Queries

 

AtScale's AI value proposition centers on making LLM-generated queries performant:

• When AI systems generate SQL queries for analytics questions, AtScale ensures those queries run fast

• Semantic metadata helps natural language tools understand which metrics and dimensions exist

• Aggregate awareness prevents AI-generated queries from scanning billions of rows unnecessarily

This approach makes sense when AI is primarily generating analytical queries that need performance optimization.

 

 AtScale makes AI queries faster. Kobai makes AI answers trustworthy. 

 

Kobai: Grounding AI in Business Context

 

Kobai's AI value proposition centers on providing structured knowledge that makes AI reliable:

• Semantic relationships for RAG/GraphRAG: Feed AI systems not just text chunks, but structured entities and how they connect.

• Traceability by design: Episteme allows users to see how AI-generated answers derive from the knowledge model.

• Governed context: AI queries inherit Unity Catalog permissions, thereby access controls enforced automatically.

• Business semantics, not just table schemas: AI understands 'supplier,' 'assembly,' 'failure mode', not just column names.

   

Why This Matters: If your AI use case is natural language BI ("Show me Q4 revenue by region"), AtScale's acceleration matters. If your AI use case is copilots for operations, compliance assistants, or context-aware agents that need to understand how the business actually works, Kobai's knowledge layer matters. Different AI problems require different semantic foundations.

 

Databricks Integration: Connected vs Native

 

AtScale + Databricks: BI Acceleration Layer

 

AtScale integrates with Databricks as a BI semantic layer:

• Connects to Databricks as a data source for aggregate building and query optimization

• Creates cubes and aggregates for accelerating BI tool queries

• Provides semantic layer between BI tools (Tableau, Power BI) and Databricks tables

• Operates as an additional platform layer that virtualizes and optimizes analytical queries

 

Kobai + Databricks: Lakehouse-Native Knowledge Layer

 

Kobai embeds knowledge semantics directly into the Databricks lakehouse:

• Built On Partner: Designed specifically for Databricks lakehouse architecture.

• Marketplace availability: Listed on Databricks Marketplace with usage-based pricing.

• Unity Catalog inheritance: Semantic queries automatically respect lakehouse governance without parallel configuration.

• Query pushdown: Semantic operations translate to SQL/Spark and execute on Databricks compute.

•  Zero data movement: Data remains in Delta Lake; semantic intelligence added without replication. 

 

Why This Matters: Both platforms can work with Databricks. The difference is architectural depth: AtScale connects to Databricks to accelerate BI. Kobai is part of Databricks lakehouse infrastructure as its semantic intelligence. For organizations committed to lakehouse consolidation, this distinction affects long-term platform alignment.

 

When to Choose Kobai Over AtScale

 

Kobai is the right choice when:

• Cross-domain understanding is the priority: You need to connect engineering, operations, supply chain, and quality data semantically, not just report on them.

• AI needs business context, not just metrics: Your copilots and agents require structured knowledge about entities and relationships to provide trustworthy answers.

• Traceability and lineage are mandatory: Compliance, quality, or regulatory requirements demand traceable connections from design through operations.

• Domain experts need semantic modeling: Business users should define what entities and relationships mean without waiting on data engineering.

• Databricks-native integration matters: You want knowledge semantics embedded in the lakehouse, not layered on top for BI acceleration.

 

When Analytics Acceleration Makes Sense

 

An analytics-focused semantic layer may be the priority when:

• BI performance is the primary concern: Dashboard query speed and aggregate management are the critical pain points.

• Metric consistency across tools: The primary goal is ensuring KPIs calculate the same way in Tableau, Power BI, and Looker.

• Analytics-centric organization: Primary users are BI analysts and dashboard consumers, not domain experts investigating operational issues.

• Established BI infrastructure: Investment in dimensional modeling and BI tooling is significant and performance optimization is the priority.

Kobai and AtScale solve different problems: AtScale optimizes the analytics workflow. Kobai enables understanding across domains. Organizations pursuing both BI excellence and AI-driven operations may eventually need both but they serve different strategic purposes.

 

Complementary Capabilities, Different Priorities

 

Analytics semantics and knowledge semantics are not competing, they're complementary. The choice depends on which problem creates more urgency for your organization today.

Ask: Is your biggest bottleneck dashboard performance and metric consistency? Or is it helping people understand complex, cross-domain relationships that don't fit neatly into BI reports?

Many enterprises eventually build both capabilities. The question is which one addresses your most pressing need first.

Strategic Perspective: If you're an asset-intensive organization (aerospace, energy, manufacturing) where operational intelligence requires understanding how physical objects, processes, and events connect across lifecycle stages, Kobai's knowledge semantics become foundational. If you're primarily optimizing reporting and analytics workflows, AtScale's acceleration delivers immediate value. Neither replaces the other as they address different layers of the semantic challenge.

 

Capability Comparison

 

 

The Right Tool for the Right Problem

 

AtScale and Kobai both add semantic intelligence to enterprise data. They solve different problems.

AtScale makes BI faster and more consistent which is critical for analytics organizations where dashboard performance and metric standardization drive value. Kobai makes complex operational reality understandable which is critical for organizations where cross-domain knowledge and AI trust determine competitive advantage.

The choice is not which platform is better. The choice is which problem matters more to your organization right now: accelerating existing analytics workflows, or enabling new forms of intelligence that require understanding how things actually connect..

Analytics semantics make queries faster. Knowledge semantics make understanding possible. Both matter but they serve different strategic purposes.

 

See how Kobai brings knowledge semantics to your Databricks lakehouse.

See Kobai Inside Databricks | Explore Customer Usecases