BRAND BOOK Databricks Inspired · v1.0

⚡ Enterprise Data + AI

The Data + AI
Platform.

Databricks Red on deep navy. Bold, enterprise-confident, data-visualization-forward. Built for the Lakehouse era — where AI meets the data warehouse.

10B+

Queries per day

10,000+

Enterprise customers

2–10×

Faster than alternatives

Foundations

Identity

Databricks

Spark-diamond logomark · DM Sans 700

Four overlapping diamond/rhombus shapes forming a stylised spark cluster — the Databricks icon, rendered in brand red on dark backgrounds. Sharp, technical, data-forward.

Voice & Tone

"Unify your data, analytics, and AI on one Lakehouse."

Data-first, proof over promise. Technical depth respected — the audience knows what a Delta table is. Scale as credibility: exabytes, milliseconds, petabytes. Confident, never hedging.

Confident Data-first Enterprise Open Source

Design Principles

🌑

Dark-first

The platform lives in dark terminals and notebooks. Light backgrounds are an exception.

🔴

Red with restraint

Red-orange is the single high-energy accent. Use only for primary CTAs and data highlights.

📊

Data viz first-class

Charts, query stats, and cluster metrics belong in hero sections — not as afterthoughts.

⬜

Sharp corners

Sharper radius than consumer products. This is enterprise infrastructure, not a lifestyle app.

Color

Primary — Databricks Red

100

200

300

400

500 ★

600

700

800

#FF3621 — Databricks Red. Confirmed from brand/press assets and site CSS. Used on CTA buttons, logo, and data emphasis. High-energy — use with restraint.

Neutrals — Dark Canvas

Canvas #0E1520

Navy #1B2631

Elevated #243342

Border #2E4053

Signature Gradient

Language / Domain Accents

SQL

#38BDF8

Python

#FBBF24

#A78BFA

Streaming

#34D399

Typography

DM Sans — Marketing

From data
to AI, faster.

DM Sans 700 · Confirmed on databricks.com marketing site. Clean, modern, confident. Technical without being cold.

JetBrains Mono — Notebooks

Python [1]

from pyspark.sql import SparkSession spark = SparkSession.builder.getOrCreate()

JetBrains Mono — used in notebook code cells and terminal output. Confirmed in Databricks workspace UI.

Type Scale

display · 64px · 700

The Data + AI Platform.

h1 · 48px · 700

Unify your Lakehouse.

h2 · 32px · 700

10B queries per day.

h3 · 22px · 500

Mosaic AI — from raw data to production.

body · 16px · 400

Databricks combines data warehousing and data lakes into one open, unified platform.

mono · 13px · 400

spark.read.format("delta").load("/mnt/lakehouse/sales")

Tokens

Border Radius — Sharp Enterprise

2px

4px

6px

8px

10px

card

Shadows

shadow-sm

0 1px 3px rgba(0,0,0,0.5)

shadow-card

0 4px 16px rgba(0,0,0,0.4)

shadow-glow-red

0 0 24px rgba(255,54,33,0.3)

Sample Data Visualization

Query Performance · ms

SQL

4.2ms

Python

8.1ms

ML train

14ms

Streaming

5.4ms

Components

Buttons

Badges

● Running ● Ready Terminated SQL Python ML Streaming

Notebook Cell — The Databricks UI Primitive

Python LakehouseDemo.ipynb ● Running

from pyspark.sql import SparkSession spark = SparkSession.builder.appName("LakehouseDemo").getOrCreate() df = spark.read.format("delta").load("/mnt/lakehouse/sales") df.filter(df.revenue > 1_000_000).groupBy("region").sum("revenue").show()

+----------+------------------+
| region| sum(revenue)|
+----------+------------------+
| APAC | 4289321000.00|
| EMEA | 6712440000.00|
|AMERICAS | 9841250000.00|
+----------+------------------+

SQL ✓ 0.8s

SELECT region, SUM(revenue) AS total_revenue, COUNT(*) AS deal_count FROM lakehouse.sales WHERE revenue > 1000000 GROUP BY region ORDER BY total_revenue DESC

Cluster Table

Cluster	Runtime	Workers	Status	Cost/hr
prod-analytics	14.3 ML	16	● Running	$4.20
dev-workspace	13.3 DBR	4	● Ready	$0.90
ml-training	14.3 ML GPU	8	Terminated	—

Form Controls

Patterns · Workspace Shell

Workspace

📊 Data Explorer

📓 Notebooks

⚡ Clusters

🤖 ML Models

🔄 Workflows

Data Explorer · lakehouse.sales

ROW COUNT

84.2M

COLUMNS

SIZE

12.4 GB

Usage · Do & Don't

✓ Do

Use red-orange for primary CTAs and key data callouts only — high-energy, use with restraint.

Lead with data — charts, query performance numbers, and cluster metrics in hero sections.

Use sharper corners than consumer products — this is enterprise infrastructure.

Include code samples — the audience validates credibility through technical depth.

✕ Don't

Don't use the red gradient on large background areas — it competes with content.

Don't soften the dark palette with pastels — the brand is authoritative, not friendly.

Don't omit code samples or technical detail — the audience reads dashboards all day.

Don't use vague AI hype without data to back it — proof over promise.

The Data + AIPlatform.