Quick platform check
Who this is for:
Architecture / Concept Overview: Quick platform check
Mosaic AI spans the full AI lifecycle inside the Databricks Data Intelligence Platform.
%%{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
DATA[Lakehouse Data] -->|Prepare| FE[Feature Engineering]
FE -->|Train| TRAIN[AutoML / Custom Training]
TRAIN -->|Track| MLF[MLflow Tracking & Registry]
MLF -->|Deploy| SERVE[Model Serving]
SERVE -->|Ground| RAG[RAG / Agents]
RAG -->|Monitor| MON[Lakehouse Monitoring]
MON -->|Feedback| DATA
DATA:::source
FE:::ingestion
TRAIN:::processing
MLF:::governance
SERVE:::serving
RAG:::processing
MON:::governance
*The Mosaic AI lifecycle: data flows from the lakehouse through training, serving, and monitoring in a continuous loop.*
%%{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
MOSAIC[Mosaic AI Platform] --> CLASSIC[Classic ML]
MOSAIC --> DL[Deep Learning]
MOSAIC --> GENAI[Generative AI]
MOSAIC --> AGENTS[AI Agents]
CLASSIC --> AUTOML[AutoML]
CLASSIC --> SKLEARN[Scikit-Learn / XGBoost]
DL --> TORCH[PyTorch / TensorFlow]
DL --> DIST[Distributed Training]
GENAI --> FM[Foundation Models]
GENAI --> FT[Fine-Tuning]
AGENTS --> AF[Agent Framework]
AGENTS --> EVAL[Agent Evaluation]
MOSAIC:::governance
CLASSIC:::processing
DL:::processing
GENAI:::serving
AGENTS:::serving
AUTOML:::ingestion
SKLEARN:::ingestion
TORCH:::storage
DIST:::storage
FM:::source
FT:::source
AF:::ingestion
EVAL:::ingestion
*Mosaic AI capability tree spanning classic ML, deep learning, generative AI, and autonomous agents.*
Key Terms
Prerequisites and Setup
- A Databricks workspace (Premium or Enterprise tier) with Unity Catalog enabled.
- Workspace admin access to enable Model Serving and AI features.
- A cluster running the latest Databricks Runtime for ML.
- Familiarity with Python and basic ML concepts.
Step-by-Step Implementation
Configuration Reference
| Parameter | Default | Description |
|---|---|---|
timeout_minutes (AutoML) | 120 | Maximum exploration time for AutoML |
max_tokens (Foundation Models) | 256 | Maximum tokens in LLM response |
scale_to_zero_enabled | true | Allow serving endpoints to scale to zero |
workload_size | Small | Compute size for serving endpoints |
experiment_path | user folder | MLflow experiment notebook path |