Study guide

7.4 AI/ML Pipelines & Applications: Cortex AI, ML Functions, Snowpark Container Services, Streamlit & Native Apps

SnowPro Advanced: Architect ARA-C01

Track your progress as you read

Finished sections get a check mark.

1 Introduction: The ARA-C01 Exam

About the SnowPro Advanced: Architect (ARA-C01) Exam

2 Chapter 1: Account Strategy & Organization Architecture

1.1 Multi-Account Strategy & Snowflake Organizations 1.2 Snowflake Parameter Hierarchy & Configuration 1.3 Database and Schema Design Architecture

3 Chapter 2: Identity, Access Control & Storage Integrations

2.1 Role-Based Access Control Architecture & Hierarchy 2.2 Secondary Roles, Object Ownership & Delegated Administration 2.3 Cloud Storage Integrations & IAM Trust Policies 2.4 Identity Federation, SSO, SCIM & Authentication Policies

4 Chapter 3: Governance, Data Privacy, Network Security & Compliance

3.1 Column-Level Masking & Row Access Policies 3.2 Object Tagging, Tag-Based Masking & Audit Governance 3.3 Encryption, Key Management & Tri-Secret Secure 3.4 Aggregation Policies, Projection Policies & External Tokenization 3.5 Network Policies, Network Rules, External Access & Private Connectivity 3.6 Snowflake Editions & Compliance Architecture (PCI DSS, PII & PHI)

5 Chapter 4: Data Modeling in Snowflake

4.1 Dimensional Modeling: Star vs. Snowflake Schemas 4.2 Data Vault 2.0 Architecture in Snowflake 4.3 Relational Constraints & Query Optimizer Integration

6 Chapter 5: Data Sharing & Collaboration Solutions

5.1 Direct Data Sharing Architecture & Security 5.2 Managed Reader Accounts for External Consumers 5.3 Snowflake Marketplace, Data Exchange & Clean Rooms

7 Chapter 6: Multi-Cloud, Multi-Region & Business Continuity

6.1 Database & Account Replication Mechanics 6.2 Disaster Recovery Architecture & Client Redirect 6.3 Cross-Cloud Auto-Fulfillment & Listing Replication 6.4 Data Protection: Time Travel, Fail-safe & Storage Costs

8 Chapter 7: Environments, Lakehouse, DevOps & AI/ML Applications

7.1 Multi-Tier Environment Strategy & Zero-Copy Cloning 7.2 Data Lakehouse Architecture & Apache Iceberg 7.3 DataOps, CI/CD & Snowflake Git Integration 7.4 AI/ML Pipelines & Applications: Cortex AI, ML Functions, Snowpark Container Services, Streamlit & Native Apps

9 Chapter 8: Data Ingestion Architecture

8.1 Bulk Ingestion Architecture: COPY INTO Optimization 8.2 Snowpipe Continuous Automated Ingestion 8.3 Snowpipe Streaming & Real-Time Ingestion 8.4 Data Unloading Architecture & Cloud Export Patterns 8.5 Schema Detection, Schema Evolution & Source-Change Architecture

10 Chapter 9: Continuous Pipelines & Stream Transformation

9.1 Change Data Capture (CDC) with Snowflake Streams 9.2 Task Orchestration, DAGs & Serverless Scheduling 9.3 Declarative Data Pipelines with Dynamic Tables 9.4 Views, Materialized Views & Transformation Table Design

11 Chapter 10: Semi-Structured Data & Ecosystem Tooling

10.1 Semi-Structured Data: VARIANT, OBJECT & ARRAY Processing 10.2 Snowflake Ecosystem Connectors, Drivers & SQL API 10.3 Snowpark, User-Defined Functions & Stored Procedures

12 Chapter 11: Virtual Warehouse Architecture & Concurrency

11.1 Virtual Warehouse Architecture, Sizing & Credit Economics 11.2 Multi-Cluster Warehouses & Concurrency Scaling Policies 11.3 Warehouse Types & Query Acceleration Service

13 Chapter 12: Query Profiling & Performance Diagnostics

12.1 Query Profile Deep-Dive & Bottleneck Analysis 12.2 Memory Spilling, Warehouse Queuing & Timeout Governance 12.3 Three-Tier Caching Architecture & Result Reuse

14 Chapter 13: Storage Optimization, Cost Governance & Observability

13.1 Micro-Partitioning & Automatic Clustering 13.2 Search Optimization Service (SOS) & Point Lookup Acceleration 13.3 Resource Monitors, Financial Governance & Usage Observability 13.4 Warehouse Monitoring, Alerts, Event Tables & Performance Explorer

7.4 AI/ML Pipelines & Applications: Cortex AI, ML Functions, Snowpark Container Services, Streamlit & Native Apps

Key Takeaways Explain This Topic AI

Snowflake Cortex AI Functions (such as AI_COMPLETE, AI_CLASSIFY, AI_FILTER, AI_EXTRACT, AI_SENTIMENT, AI_EMBED) run LLMs inside Snowflake's service perimeter from SQL or Python; callers need the USE AI FUNCTIONS privilege plus the CORTEX_USER or AI_FUNCTIONS_USER database role.

ML Functions (Forecasting, Anomaly Detection, Classification, Top Insights) let analysts train and use models from SQL without writing ML code; model instances incur storage and compute and require AUTOCOMMIT to be enabled.

Snowpark Container Services runs OCI container images on Snowflake-managed compute pools (CPU or GPU) for custom runtimes, model training, and model serving, using network policies for ingress, external access integrations for egress, and event tables for logs.

Streamlit in Snowflake stores app code in a Snowflake object governed by RBAC and runs on a warehouse or container runtime, so data never leaves Snowflake to power an interactive app.

The Native App Framework packages data, logic, and Streamlit UIs into an application package (manifest plus setup script, with versions and patches) that providers distribute through free or paid listings on the Marketplace or privately.

Last updated: September 2026

Why AI/ML Belongs in an Architecture Exam

Objective 2.3 of the blueprint includes "Outline basic AI/ML pipelines and applications" and names Snowpark Container Services , Snowflake ML functions , Cortex LLM functions , Streamlit , and the Snowflake Native App Framework . The exam does not expect data-science depth; it expects you to pick the right building block for a scenario and to understand its security, cost, and operational implications.

  • Need Building block Skill level required
  • Summarize, classify, extract from, or translate text/images with an LLM in SQL Cortex AI Functions SQL analyst
  • Forecast a metric, detect anomalies, classify rows, or find top drivers ML Functions SQL analyst

Build, train, register, and serve custom models Snowflake ML (Snowpark ML, Model Registry, Feature Store) Data scientist

Run arbitrary containers, custom runtimes, GPUs, or long-running services Snowpark Container Services (SPCS) Engineer

  • Put an interactive UI on Snowflake data Streamlit in Snowflake Python developer
  • Distribute data plus logic to other accounts Native App Framework Provider engineering team
  • Snowflake Cortex AI Functions

Cortex AI Functions run industry-leading LLMs (from providers such as OpenAI, Anthropic, Meta, Mistral AI, and DeepSeek) inside the Snowflake service perimeter and are called like any SQL function (they are also available in Python): -- Classify and score support tickets without moving data out of Snowflake

  • SELECT ticket_id,
  • AI_SENTIMENT(body) AS sentiment,
  • AI_CLASSIFY(body, ['billing', 'outage', 'feature request']) AS category,
  • AI_EXTRACT(body, {'product': 'Which product is mentioned?'}) AS extracted
  • FROM support.tickets
  • WHERE AI_FILTER(PROMPT('Is this ticket about a data outage? {0}', body));

Architect considerations:

Access control: the role needs the account-level USE AI FUNCTIONS privilege and one of the CORTEX_USER or AI_FUNCTIONS_USER database roles.

Data privacy: models run inside Snowflake's perimeter, and Snowflake does not use customer data to train models offered to other customers.

Availability and cost: functions and models are available in select regions (cross-region inference can be enabled with CORTEX_ENABLED_CROSS_REGION ); usage is billed in credits and appears in usage views under AI service types. Some functions are preview features — check status before production use.

Related services: Cortex Search (hybrid search for RAG), Cortex Analyst (natural-language questions over semantic models), and Cortex Agents build on the same governed foundation.

ML Functions

ML Functions give SQL users out-of-the-box machine learning:

Forecasting and Anomaly Detection for time-series metrics.

Classification to sort rows into classes.

Top Insights to find dimensions that drive unexpected changes in a metric. CREATE SNOWFLAKE.ML.FORECAST daily_sales_model(

  • INPUT_DATA => SYSTEM$REFERENCE('VIEW', 'analytics.daily_sales_v'),
  • TIMESTAMP_COLNAME => 'sale_date',
  • TARGET_COLNAME => 'revenue');

CALL daily_sales_model!FORECAST(FORECASTING_PERIODS => 30);

Models are schema-level objects; training and prediction use warehouse compute, and model artifacts incur storage (delete obsolete models). AUTOCOMMIT must be enabled in the session.

Snowflake ML and Snowpark for Custom Models

For custom models, data scientists use Snowpark (Python DataFrames executed in Snowflake) and Snowflake ML — modeling APIs based on popular frameworks, a Feature Store , and a Model Registry for versioning and deploying models. Training can run on Snowpark-optimized warehouses (high memory) or on SPCS compute pools (including GPUs); registered models can serve predictions from SQL or from a service.

Snowpark Container Services (SPCS)

SPCS is a fully managed container platform inside Snowflake (generally available in AWS, Azure, and Google Cloud commercial regions, with some exceptions). You package an application and its dependencies as an OCI image , push it to an image repository in Snowflake, and run it on a compute pool . CREATE COMPUTE POOL ml_gpu_pool

  • MIN_NODES = 1 MAX_NODES = 2
  • INSTANCE_FAMILY = GPU_NV_S;

CREATE IMAGE REPOSITORY ml.images.repo;

  • CREATE SERVICE ml.serving.fraud_model_svc
  • IN COMPUTE POOL ml_gpu_pool
  • FROM @ml.specs.stage SPECIFICATION_FILE = 'fraud_svc.yaml'
  • EXTERNAL_ACCESS_INTEGRATIONS = (model_hub_access);

What it is for:

Custom runtimes and libraries that UDFs cannot run; GPU workloads such as model training and LLM or model serving.

Long-running services (APIs, web apps) and job services (batch jobs run with EXECUTE JOB SERVICE ).

How it integrates with Snowflake governance:

Services can run SQL in a virtual warehouse and read staged files.

Network ingress is governed by network policies; egress requires external access integrations .

RBAC controls which roles can use a service and how services talk to each other.

Logs, metrics, and events go to event tables .

Reader accounts cannot create services, and cross-cloud auto-fulfillment of Native Apps with SPCS is supported only on AWS and Azure.

Streamlit in Snowflake

Streamlit in Snowflake lets developers build and share Python web apps on Snowflake data without moving the data or the code elsewhere:

The app's source code and environment are stored in a Snowflake object governed by RBAC .

Snowflake manages compute; apps run on a warehouse runtime or a container runtime .

Apps work with Snowpark, UDFs, stored procedures, and Native Apps, and can be created in Snowsight, with SQL, or with the Snowflake CLI.

Snowflake Native App Framework

The Native App Framework lets a provider share data and application logic with consumers :

An application package holds the data content, logic (stored procedures, functions, Streamlit UI, optionally containers), a manifest file , and a setup script that runs when a consumer installs or upgrades the app.

Providers release versions and patches , test from a single account, and collect logs and events for troubleshooting.

Apps are distributed through listings — free or paid, on the Snowflake Marketplace or privately to specific accounts — and can use Cross-Cloud Auto-Fulfillment to reach other regions.

Consumers install the app into their own account, so their data never has to be sent to the provider.

An End-to-End AI/ML Pipeline in Snowflake Ingest (Snowpipe / Streaming) → Features (dynamic tables, Feature Store)

  • → Train (Snowpark ML on Snowpark-optimized WH or SPCS GPU pool)
  • → Register (Model Registry) → Score (SQL/Python inference, Cortex AI Functions)
  • → Serve UI (Streamlit) → Distribute (Native App via listing)

Every step stays inside Snowflake's security perimeter, so the same RBAC, masking, row access policies, and auditing (Chapters 2–3) govern ML data as govern analytics.

Loading diagram... AI/ML Building Blocks and Their Place in a Pipeline