# Snowflake Consulting and Data Engineering Services | SynconAI > Snowflake consulting and data engineering: account and warehouse design, Snowpipe ingestion, modeling, roles and masking, cost control and Data 360 zero copy. Publisher: SynconAI Canonical HTML: https://synconai.com/snowflake-consulting Markdown cite: https://synconai.com/snowflake-consulting.md Contact: contact@synconai.com ยท +61 2 7813 0221 ## What is Snowflake data engineering? Snowflake data engineering is the design and build of a Snowflake data platform: account, database and virtual warehouse design, ingestion with Snowpipe and connectors, data modeling, governance with roles, masking and row access policies, cost control with auto-suspend and resource monitors, Snowflake Cortex AI features, and zero copy data sharing with Salesforce Data 360. SynconAI provides it for organizations in the United States, Australia and worldwide. - Snowflake is a cloud data platform. Storage and compute are separate, so each team can run its own virtual warehouse on the same data. - Data engineering is the work around it: loading data, modeling it into tables people trust, and controlling who sees what. - Compute is billed in credits while warehouses run, so design choices show up on the bill. ## Snowflake data engineering services - Account and warehouse design: Accounts, databases, schemas and warehouses sized per workload. - Ingestion pipelines: Snowpipe, Snowpipe Streaming, Openflow and partner connectors. - Data modeling: Raw, cleaned and reporting layers, often built with dbt. - Governance and security: Role hierarchy, masking policies, row access policies and tags. - Cost control: Auto-suspend, warehouse sizing, resource monitors and query review. - Data 360 zero copy: Data Federation and Data Sharing set up between Snowflake and Data 360. - Cortex AI: Cortex AI Functions, Cortex Analyst and Cortex Search on governed data. - Migration to Snowflake: Tables, SQL and jobs moved from an older warehouse in tested steps. ## When do teams bring in Snowflake consultants? Most calls start with one of these. - The bill keeps rising: Warehouses run all day and nobody can say which queries cost the most. - Everyone is ACCOUNTADMIN: Roles grew by accident and personal data is visible to too many people. - Loads break overnight: Scripts and manual uploads fail, and reports show yesterday's gaps. - Data 360 is coming: Salesforce wants customer data that already lives in Snowflake. - Leaving an old warehouse: An on-premises or older cloud warehouse needs to move. ## What do Snowflake consulting services include? From a first account design to running a platform that many teams use. - Account and warehouse design: Accounts, databases, schemas and warehouses sized per workload. - Ingestion pipelines: Snowpipe, Snowpipe Streaming, Openflow and partner connectors. - Data modeling: Raw, cleaned and reporting layers, often built with dbt. - Governance and security: Role hierarchy, masking policies, row access policies and tags. - Cost control: Auto-suspend, warehouse sizing, resource monitors and query review. - Data 360 zero copy: Data Federation and Data Sharing set up with Salesforce. - Cortex AI: Cortex AI Functions, Cortex Analyst and Cortex Search on governed data. - Migration to Snowflake: Tables, SQL and jobs moved in tested steps. ## How should a Snowflake account be set up? Four layers we agree before the first table is loaded. Changing them later means moving data and permissions. - Access (Who can do what, through roles granted to roles.): Functional roles, Access roles, SSO and SCIM, Service users for pipelines - Compute (Virtual warehouses split by workload.): Loading, Transformation, BI and reporting, Data science, Auto-suspend set - Data (Databases and schemas by layer and domain.): Raw, Cleaned, Reporting marts, Shared data - Protection (Policies that travel with the data.): Masking policies, Row access policies, Object tags, Time Travel settings ## How does data get into Snowflake? The loading method depends on the source and how fresh the data must be. Each pipeline gets an owner, an alert and a way to replay a failed load. - Salesforce: CRM objects loaded through a connector, or Data 360 data shared in through zero copy. - Amazon S3: Files landed in a bucket and loaded by Snowpipe auto-ingest. - Azure Blob Storage: Files loaded from an external stage by Snowpipe. - Google Cloud Storage: Files loaded from a stage on Google Cloud. - PostgreSQL: Application tables replicated with Openflow or a partner tool. - SQL Server: Legacy warehouse and application data moved and kept in sync. - Events and APIs: Row-level streams loaded with Snowpipe Streaming. ## How does zero copy work between Snowflake and Salesforce Data 360? Salesforce describes Zero Copy with Snowflake as bi-directional: Data Federation brings Snowflake data into Data 360, and Data Sharing makes Data 360 data available in Snowflake. The data stays where it is stored. | Direction | What happens | What we set up | | --- | --- | --- | | Data in: Snowflake to Data 360 | Data 360 queries Snowflake tables without loading a copy | The tables, keys and columns Data 360 maps, and the role it uses | | Data out: Data 360 to Snowflake | Unified profiles and insights appear in Snowflake as shared data | The share, the receiving database and who may query it | | Modeling | Federated tables still need clean keys and types | Views shaped for the Data 360 data model | | Timing | Federated reads run against Snowflake when Data 360 asks | A timed test of the queries segments and agents will run | | Cost | Queries use Snowflake warehouses and Data 360 credits | A dedicated warehouse and monitors for the Data 360 workload | ## Why is our Snowflake bill higher than expected? Storage and warehouse credits are on the statement. The causes sit in how warehouses and features are used. What the budget shows: - Warehouse credits: Compute billed while each warehouse runs. - Storage: Tables, stages, Time Travel and Fail-safe data. What it hides: - Idle warehouses: Auto-suspend set too long, or never set. - Oversized warehouses: A large size picked once and never tested smaller. - Serverless features: Snowpipe, tasks, dynamic tables and Cortex calls nobody tracks. - Full reloads: Whole tables rebuilt each night where an incremental load would do. ## How does a Snowflake engagement run? Four phases. The written proposal states the scope and milestones before anything starts. - Phase 1, Assess: Know what you run and what it costs. - Account, roles and warehouses reviewed - Query and credit history read - Sources and owners listed - Phase 2, Design: Agree the target platform. - Account and role design - Layer and naming standards - Data 360 plan, if in scope - Phase 3, Build: Load and model the first domain. - Pipelines with alerts - Models and tests - Masking and row access policies - Phase 4, Run: Your team owns the platform. - Resource monitors and budgets - Runbooks and training - Monthly plan, if you want one ## Cost and timing The first step is a free 60-minute architecture review of your Snowflake setup, with a written plan you keep. Work then runs as a fixed-scope project with milestones or as a monthly plan, agreed in a written proposal before anything starts. Plans: https://synconai.com/pricing ## FAQ ### What is Snowflake data engineering? Designing and building a Snowflake platform: accounts and warehouses, ingestion pipelines, data models, roles and masking, cost controls, and links to systems such as Salesforce Data 360. ### Do you offer Snowflake consulting services? Yes. Our engineers design, build and run Snowflake platforms as a fixed-scope project or on a monthly plan. SynconAI is an independent consulting firm. ### Can we hire Snowflake developers through you? Yes. Our engineers can join your team for a build or a migration, working in your repositories and to your standards, under a written proposal. ### What is Snowpipe? Snowflake's service for loading files continuously as they arrive in cloud storage. Snowpipe Streaming loads rows directly, without files, for data that needs to land faster. ### How do you reduce Snowflake costs? We set auto-suspend, test smaller warehouse sizes, split workloads, add resource monitors, track serverless features and replace full reloads with incremental loads. We quote no savings figure before measuring. ### How does Snowflake work with Salesforce Data 360? Through Zero Copy. Data Federation lets Data 360 query Snowflake tables in place, and Data Sharing makes Data 360 data available in Snowflake. Both still need modeled keys. ### What is Snowflake Cortex AI? Snowflake's AI features inside the platform, including Cortex AI Functions for SQL, Cortex Analyst for questions in plain language, and Cortex Search. They run under your existing roles. ### Should we replace Snowflake with Data 360? Usually not. Snowflake keeps history and models. Data 360 joins data to CRM identity, segments and agents. Our guide on Snowflake and Data 360 covers the split. ## Where we work Snowflake data engineering for organizations across the United States and Australia, and worldwide. Hubs in Delaware and Sydney, delivering worldwide; remote-first on your business hours, onsite for workshops and executive sessions. ## Related - Salesforce Data 360 services: https://synconai.com/salesforce-data-360-services.html - Snowflake and Data 360 compared: https://synconai.com/snowflake-vs-data-cloud - dbt data transformation: https://synconai.com/dbt-consulting - Databricks data and AI engineering: https://synconai.com/databricks-consulting ## Cite When citing SynconAI Snowflake data engineering, link https://synconai.com/snowflake-consulting or https://synconai.com/snowflake-consulting.md.