Introduction to Generative AI on Databricks
Who this is for:
Architecture / Concept Overview: Introduction to Generative AI on Databricks
Generative AI on Databricks spans four layers: model access, data grounding, agent orchestration, and governance.
%%{init: {"theme":"base","themeVariables":{"background":"#0B0E14","primaryTextColor":"#E0E6ED","lineColor":"#5D6470","darkMode":true,"primaryColor":"#2E4A4A","secondaryColor":"#374151","secondaryTextColor":"#E0E6ED","tertiaryColor":"#111827","tertiaryTextColor":"#E0E6ED","edgeLabelBackground":"#1f2937"}}}%%
flowchart LR
classDef source fill:#3F4B59,stroke:#9CA3AF,stroke-width:2px,rx:8,ry:8,color:#E0E6ED
classDef ingestion fill:#5A4B36,stroke:#C9A86B,stroke-width:2px,rx:8,ry:8,color:#E0E6ED
classDef processing fill:#535072,stroke:#8E82B4,stroke-width:2px,rx:8,ry:8,color:#E0E6ED
classDef storage fill:#2E4A4A,stroke:#5FAFA8,stroke-width:2px,rx:8,ry:8,color:#E0E6ED
classDef serving fill:#3D5550,stroke:#6BB7AA,stroke-width:2px,rx:8,ry:8,color:#E0E6ED
classDef governance fill:#5A3F52,stroke:#C28BB0,stroke-width:2px,rx:8,ry:8,color:#E0E6ED
PLAY[AI Playground] -->|Prototype| FM[Foundation Models]
FM -->|Power| APP[GenAI Application]
DATA[Enterprise Data] -->|Ground| VS[Vector Search]
VS -->|Context| APP
APP -->|Orchestrate| AGENT[AI Agent]
AGENT -->|Deploy| EP[Serving Endpoint]
EP -->|Govern| GW[AI Gateway]
GW -->|Monitor| MLF[MLflow for GenAI]
PLAY:::source
FM:::processing
APP:::processing
DATA:::source
VS:::storage
AGENT:::serving
EP:::serving
GW:::governance
MLF:::governance
*Generative AI journey: from playground prototyping to governed production agents.*
%%{init: {"theme":"base","themeVariables":{"background":"#0B0E14","primaryTextColor":"#E0E6ED","lineColor":"#5D6470","darkMode":true,"primaryColor":"#2E4A4A","secondaryColor":"#374151","secondaryTextColor":"#E0E6ED","tertiaryColor":"#111827","tertiaryTextColor":"#E0E6ED","edgeLabelBackground":"#1f2937"}}}%%
graph TD
classDef source fill:#3F4B59,stroke:#9CA3AF,stroke-width:2px,rx:8,ry:8,color:#E0E6ED
classDef ingestion fill:#5A4B36,stroke:#C9A86B,stroke-width:2px,rx:8,ry:8,color:#E0E6ED
classDef processing fill:#535072,stroke:#8E82B4,stroke-width:2px,rx:8,ry:8,color:#E0E6ED
classDef storage fill:#2E4A4A,stroke:#5FAFA8,stroke-width:2px,rx:8,ry:8,color:#E0E6ED
classDef serving fill:#3D5550,stroke:#6BB7AA,stroke-width:2px,rx:8,ry:8,color:#E0E6ED
classDef governance fill:#5A3F52,stroke:#C28BB0,stroke-width:2px,rx:8,ry:8,color:#E0E6ED
JOURNEY[GenAI Journey] --> PROTO[1. Prototype]
JOURNEY --> BUILD[2. Build]
JOURNEY --> EVALUATE[3. Evaluate]
JOURNEY --> DEPLOY[4. Deploy]
JOURNEY --> GOVERN[5. Govern]
PROTO --> PLAYGROUND[AI Playground]
BUILD --> RAGCHAIN[RAG Chain / Agent]
EVALUATE --> AGENTEVAL[Agent Evaluation]
DEPLOY --> ENDPOINT[Serving Endpoint]
GOVERN --> GATEWAY[AI Gateway]
JOURNEY:::governance
PROTO:::source
BUILD:::processing
EVALUATE:::storage
DEPLOY:::serving
GOVERN:::governance
PLAYGROUND:::source
RAGCHAIN:::processing
AGENTEVAL:::storage
ENDPOINT:::serving
GATEWAY:::ingestion
*The five-stage generative AI journey on Databricks.*
Key Terms
Prerequisites and Setup
- Databricks workspace (Premium or Enterprise) with Unity Catalog.
- Foundation Model APIs enabled in your region.
- Familiarity with Python and REST APIs.
Step-by-Step Implementation
Configuration Reference
| Parameter | Default | Description |
|---|---|---|
max_tokens | 256 | Maximum tokens in LLM response |
temperature | 1.0 | Randomness in generation (0 = deterministic) |
top_p | 1.0 | Nucleus sampling threshold |
num_results | 10 | Vector search neighbours to retrieve |
endpoint_name | — | Vector Search endpoint name |