Logiciel Solutions Contact Us
Success Stories Tech News Investors Contact Us
whitepaper
7 min read

Lakehouse AI: Databricks and Snowflake Cortex Implementation Guide.

Most companies that want GenAI on their own data already pay for Snowflake, Databricks or both, and moving that data to a separate AI stack adds cost, copies and governance gaps. Snowflake Cortex AI and Databricks Agent Bricks, Genie and AI Functions now cover the same capability classes: SQL LLM functions, hybrid search, text-to-SQL, agent runtimes and business-user assistants.

This guide maps them side by side using official documentation checked in November 2026, covering architecture, governance, pricing units, Apache Iceberg interoperability, four use-case patterns and a weighted scorecard. It is a vendor-neutral reading of public docs and list prices, not a benchmark of either platform, and negotiated rates, regions and feature status will differ.

In depth

Same Building Blocks, Different Places for the Developer.

01

Team shape should decide the platform before features do.

Both platforms bring the model to governed data instead of exporting data to a model, and they differ mainly in where they put the developer. Snowflake keeps most work inside SQL and managed services, so an analyst can classify a million support tickets with one AI_CLASSIFY query and never open a Python environment. Databricks documents Agent Bricks as a code-first platform for developers who bring frameworks such as LangGraph or the OpenAI Agents SDK and choose any model, with low-code needs moving to Genie Agents and Genie One. SQL-first analytics teams get to production faster on Cortex AI, while teams with ML engineers get more room on Databricks.

In shortSQL-first analytics teams get to production faster o…
02

Retrieval permissions are the most common governance gap.

On Snowflake, AI access is granted like any other privilege, CoWork respects existing row access and masking policies, and June 2026 announcements added agent identity and data movement policies that can stop an agent exporting data the user could export. On Databricks, Unity Catalog governs data, tools, models and agents, and Unity Gateway applies guardrails and rate limits to model and MCP calls. Databricks AI Search, however, does not support row- or column-level permissions, so filtering has to happen in the application layer. Build ACL filters into every retrieval call and test Cortex Search results against your policies before go-live.

In shortBuild ACL filters into every retrieval call and test…
03

Per-token rates rarely decide total cost.

Snowflake bills AI features in AI Credits at $2.00 with global routing or $2.20 with regional routing, while Databricks bills DBUs multiplied by your SKU price, with a 10% uplift on residency-marked models. The bigger drivers sit elsewhere. Cortex Search bills per GB-month while resumed, even with zero queries. AI_CLASSIFY counts every label and example as input tokens on every row. Agent costs add up across every tool call, and Snowflake warehouses larger than MEDIUM add cost without speeding inference. Price each workload with a 1% sample before you scale.

In shortPrice each workload with a 1% sample before you scale
04

Apache Iceberg lets one copy of data serve both platforms.

Unity Catalog managed Iceberg tables became generally available in May 2026, and external engines including Snowflake can read and write them through the Unity Catalog Iceberg REST catalog. On the Snowflake side, a catalog integration and a catalog-linked database discover and sync remote namespaces and tables. Writes are on by default, and dropping a table with writes enabled deletes it in both systems. Catalog sync moves metadata, not every masking or row filter rule, so policies must be reapplied on each side. Assign one owning catalog per table and mount the other side read-only.

In shortAssign one owning catalog per table and mount the ot…
The detail

Three Decisions That Settle Most Lakehouse AI Projects Early.

Most of the risk in a lakehouse AI program is decided before any model is called. Gartner ties expected agentic AI cancellations to cost, unclear value and weak risk controls. Three decisions address those directly, and each one leaves a named artifact the team can check later.

Zone · 01

Which platform owns each workload

Score every workload on eight weighted criteria, from data gravity at 20% down to business-user rollout at 5%, rating each platform from 1 to 5. A weighted gap of 10 points or more is a clear choice. A smaller gap means defaulting to the platform that holds the owning catalog, or using both through Iceberg. The artifact is a completed scorecard per workload, revisited every six months.

Zone · 02

How quality will be measured

Build the evaluation set before the prototype, with 100 to 300 real questions or records and their expected outputs. Use MLflow 3 scorers and LLM judges on Databricks, or Cortex AI Evaluation and AI Observability on Snowflake. Promote only when quality, latency and cost per task meet the targets agreed at the start. The artifact is a versioned evaluation set with pass thresholds written down.

Zone · 03

What spend is allowed

Set budgets and alerts on day one rather than after the first invoice. Snowflake exposes per-query credits and tokens in CORTEXFUNCTIONSQUERYUSAGEHISTORY, and Databricks records AI Function inference in system.billing.usage. Put a cost-per-successful-task metric next to every quality metric, because a cheap model that fails half the time costs more. The artifact is a budget and quota configuration live before the pilot.

By the numbers

The figures that make it a board-level conversation.

40%+
of agentic AI projects expected to be canceled by the end of 2027, citing cost, unclear value and weak risk controls
40%
of large-company respondents scaling AI agents in at least one function, up from 27% a year earlier
30 times
or more, the spread in output-token rates across models on the Databricks DBU price list
Inside the report

What you'll take away.

01

Step 1 - Pick one workload with a measurable baseline and inventory its data

Choose ticket triage time, analyst hours per report or a similar baseline, then list source tables, owning catalog, sensitivity tags and the roles that should see answers.

02

Step 2 - Score the workload and confirm feature status in your region

Use the weighted scorecard rather than forcing every workload onto one platform. Check whether each function is GA or Preview in your region and set cross-region routing deliberately.

03

Step 3 - Build the evaluation set, then prototype with managed components

Collect 100 to 300 real examples with expected outputs first. Then use AI Functions, Cortex Search or AI Search, and Cortex Analyst or Genie before writing custom orchestration.

04

Step 4 - Add guardrails and budgets, then pilot with 20 to 50 users

Apply least-privilege roles, PII redaction, rate limits and per-user quotas. Run a four-week pilot, review failures weekly and scale only through a review gate tied to NIST AI RMF.

Questions

Frequently asked.

What is Snowflake Cortex AI?

Snowflake Cortex AI is Snowflake's built-in suite for generative AI on data stored in Snowflake. It includes SQL AI Functions such as AICOMPLETE and AICLASSIFY, Cortex Search for hybrid retrieval, Cortex Analyst for text-to-SQL, Cortex Agents for orchestration and Snowflake CoWork for business users.

Models run inside the Snowflake service perimeter, and access uses standard Snowflake roles and privileges.

How does Snowflake Cortex AI pricing work?

Since April 2026, Cortex features bill in AI Credits priced at $2.00 with global routing or $2.20 with regional routing, separate from Platform Credits. AI Functions bill per million tokens by model.

Cortex Search bills per GB-month of indexed data plus embedding tokens. Agents bill orchestration tokens plus the cost of every tool they call. Warehouses running the surrounding SQL still use Platform Credits.

Is Snowflake Cortex better than Databricks for AI agents?

It depends on the agent. Cortex Agents suit configuration-driven agents that combine Cortex Analyst, Cortex Search and stored procedures inside Snowflake governance. Databricks Agent Bricks suits custom agents built with frameworks like LangGraph, with persistent memory, code execution and model choice through Unity Gateway.

Many enterprises use Snowflake for analytics agents and Databricks for engineering-heavy agents.

Can Snowflake and Databricks share data through Apache Iceberg tables?

Yes. Unity Catalog exposes managed Iceberg tables through an Iceberg REST catalog that external engines, including Snowflake, can read and write. Snowflake connects through a catalog integration and a catalog-linked database that syncs Unity Catalog namespaces and tables.

Assign one owning catalog per table, and note that drops propagate to the remote catalog when writes are enabled.

How do I control the cost of Cortex AI and Databricks AI Functions?

Measure spend per query, then cap it before the pilot. Use Snowflake's CORTEXFUNCTIONSQUERYUSAGEHISTORY view or Databricks' system.billing.usage table. Keep Snowflake warehouses at MEDIUM or smaller, suspend idle search services, pre-filter rows with SQL and keep label sets short.

Route simple tasks to small models, and set budgets and per-user quotas from day one.

What do I get in this whitepaper?

You get a side-by-side component map of Snowflake Cortex AI and Databricks, a governance comparison, a pricing breakdown of AI Credits versus DBUs, Iceberg rules for running both platforms, four use-case patterns, a 10-step implementation plan and an eight-criterion weighted selection scorecard.

Every product claim is tied to official documentation, pricing pages or press releases checked in November 2026.

How far does this evidence reach?

It reaches as far as public vendor documentation and list prices from November 2026, plus market figures from Gartner and McKinsey. It is not a performance benchmark of either platform and it reports no customer results.

Negotiated rates, regions, editions and feature status will differ, and both vendors renamed or re-scoped major AI products in 2026, so check each source before you build.

Get the whitepaper

Have it emailed to you.

Drop your details and we'll send Lakehouse AI: Databricks and Snowflake Cortex Implementation Guide straight to your inbox - no spam, unsubscribe anytime.

Download whitepaper
Next step

Score your first lakehouse AI workload before you pick a platform.

Bring one candidate workload, the tables behind it and the platform or platforms you already run. We will work through the scorecard, the evaluation set and the cost guardrails with you in a single working session. SECTION 7 - FAQ - 5 to 8 questions

Book a workload review