saasinator.ai
PLATFORM · SNOWFLAKE · CORTEX AI

Cortex AI is the per-call upsell
sitting on top of credits.

Cortex Complete, Cortex Analyst, Document AI — every call charged against the warehouse credit you already bought. You do not have to open by cancelling any of it. Start by building the same layer beside it.

  • You own the pipelines and the evals
  • No per-call credit premium
  • Runs where you choose
Start an ignite buildSee the Snowflake replacement
The estatereplaceable
  • Cortex Complete · LLM FunctionsLLM pipelines
  • Cortex AnalystAnalytics agent
  • Document AIExtraction pipeline
  • Cortex SearchRetrieval index
  • Cortex AgentsModel layer
  • Snowflake Native AppAudit-trail ledger
Cortex Complete · Cortex Analyst · Document AI · Cortex Searchyours on commit one
Why this domain exists

You already pay to store the data.
Now you pay again to ask it a question.

The capability is real. Its commercial shape is: the AI premium is metered in the same credits as the warehouse, so the bills move together and neither negotiates alone.

Cortex AI is Snowflake's in-warehouse AI service — Cortex Complete and LLM Functions for model calls, Cortex Analyst for text-to-SQL, Document AI for unstructured extraction, Cortex Search for retrieval, Cortex Agents for orchestration. Every call is priced in credits on top of the consumption bill the warehouse already runs, so the AI premium compounds the data bill.

Priced per credit, per call

Every model call is metered in credits on top of the consumption bill the warehouse already runs. Asking more questions grows the bill whether or not the answers are worth more.

Stacked on the data bill you already pay

Cortex Analyst, Document AI and Cortex Search each add a premium over the same warehouse. Small on their own; together a second data budget.

Locked to the vendor's model selection

Cortex ships Snowflake's model list. Swapping a model becomes a vendor conversation rather than an engineering one.

Double meterSnowflake's published pricing model. Both meters are theirs, not ours.
Per credit, then per callThe AI premium is metered in the same credits as the warehouse.
A question asked of the warehouse you already pay for
The consumption bill the warehouse already runs
Every model call metered in credits on top of it
STAGE 01Ignite

Build the layer beside the one you meter.

Model calls, retrieval and agent orchestration of your own, running against the same warehouse, next to Cortex. Nothing retired, nothing migrated — the Snowflake contract is untouched on day one.

LLM workflows

Model calls against your warehouse data on the model layer you choose, with prompts and an eval suite your team can read and re-run.

Runs beside Cortex Complete and LLM Functions

Semantic retrieval

Vector search, semantic retrieval and hybrid ranking built on pgvector, Qdrant or Weaviate against the warehouse you already run.

Runs beside Cortex Search

Agent platform

Multi-step agents reading your warehouse, calling tools and writing back when a human signs off, on the orchestration framework you pick.

Runs beside Cortex Agents and Snowflake Native App

Model choice

Pick the foundation model that fits the workload and swap it without a vendor migration. Cortex ships Snowflake's model selection; you ship yours.

Runs beside Cortex AI

Start an ignite build
STAGE 02Reforge

What your analysts already built on Cortex, rebuilt as software you own.

The semantic layer behind text-to-SQL, the extraction templates your team trained, and the in-warehouse applications your platform team ships. Same logic, modern substrate, and the credit meter stops.

IN — what you run today
  • Cortex Complete · LLM Functions
  • Cortex Analyst
  • Document AI
  • Cortex Search
  • Cortex Agents
  • Snowflake Native App
SAIF

the saasinator AI Factory — glass-walled delivery

  • Brief
  • Build
  • Evals
  • Deploy
  • Transfer
OUT — what you own afterwards
  • LLM pipelines
  • Analytics agent
  • Extraction pipeline
  • Retrieval index
  • Model layer
  • Audit-trail ledger

Product names are Snowflake's own. What comes out the other side is yours — source, models, prompts, evals and pipeline, transferred on commit one.

Cortex Analyst rebuild

The text-to-SQL surface and the semantic layer your analysts have already tuned, rebuilt against an eval suite your team reads and re-runs.

Cortex Analyst

Document AI rebuild

PDF, contract and form extraction rebuilt on open document-AI pipelines running on infrastructure you operate, with your own templates carried across.

Document AI

Native App rebuild

The in-warehouse applications your platform team already ships, rebuilt as software that runs where you choose rather than only inside the warehouse.

Snowflake Native App

Coverage

Four surfaces. The same pattern in each.

Wherever query volume is highest is where the per-call premium bites hardest. These are the surfaces we rebuild, and what sits inside each.

LLM workflows
  • Model calls against your warehouse on the layer you choose
  • Prompts and eval suite your team reads and re-runs
  • Model choice recorded per workload
Analytics agent
  • Natural-language analytics over your warehouse
  • Tuned to your semantic layer, not a vendor default
  • Answers traceable to the query that produced them
Document extraction
  • PDF, contract and form extraction on pipelines you operate
  • Your own templates and training data carried across
  • Output written back into the warehouse you already run
Retrieval
  • Vector search and hybrid ranking on pgvector, Qdrant or Weaviate
  • Indexes built against your warehouse, held on your infrastructure
  • No per-index fee for retrieving your own data
Functional areas this domain touches
Data & AnalyticsTechnology
Plan a rebuild
STAGE 03Liberate

The premium comes out. The workflows stay — as software you own.

Liberation is earned, not sold. By the time the Cortex line comes off the credit bill, you have watched us build the layer and run the evals that gate it.

The licence ledger

Illustrative

The commercial shape of an AI premium, as a buyer reads it

Basis of charge
Per call, metered in the same credits as the warehouse. Query volume is the meter, not the value of the answer.
Discount condition
Conditional on the consumption commitment already made for the warehouse, not on the AI layer standing on its own merits.
Premium structure
Each service — analytics, extraction, retrieval, orchestration — prices separately on top of the same consumption bill.
On exit
Semantic layer, extraction templates and retrieval indexes sit inside the vendor's warehouse — not in an asset you hold.
Cost of staying————

Shape only — the direction of travel, not a quantity. Your own curve comes from your own consumption commitment and your own query volume.

Illustrative. This is our reading of a commercial pattern, not a quotation from any Snowflake agreement — no contract text, no clause references, no figures.

Cortex Complete replacement

Model calls run on the layer you choose, on infrastructure you operate, so the per-call line comes off the credit bill entirely.

Cortex Complete · LLM Functions

Cortex Search replacement

Retrieval indexes held on your own infrastructure, so searching your own warehouse is not charged per index.

Cortex Search

Cortex Agents replacement

Multi-step orchestration owned outright, retired into only once the replacement is carrying the work.

Cortex Agents

Plan a liberation
How it is built

Glass-walled from brief to transfer. Nothing behind a black box.

SAIF is our delivery method and it runs in the open. You watch the build as it happens, read the evals that gate every release, and keep every artefact — including the ones that record what did not work.

  1. 01

    Brief

    One workflow, scoped against your own data and your own renewal position.

  2. 02

    Build

    Agentic delivery against your systems, visible while it runs.

  3. 03

    Evals

    Every release gated on tests you can read and re-run yourself.

  4. 04

    Deploy

    Into infrastructure you control, alongside the system it stands beside.

  5. 05

    Transfer

    Your team runs it. We do not leave until they can.

You own it from commit one

Source, models, prompts, evals and pipeline. Not a licence to use what we built — the asset itself.

Two weeks to a working build

A working build against your own systems in two weeks. Fixed scope, fixed bill.

It runs beside the Cortex service first

Nothing is retired on faith. The Cortex service goes when the replacement is carrying the work.

See the methodology
Proof

What we will put in writing.

100%
IP transferred on commit one
0
per-call Cortex credit consumption
2 weeks
to a working build · fixed scope, fixed bill
Our commitment

Every engagement starts with a scoped working build against your own systems. If it doesn't convince you, you pay nothing — and you keep the code either way.

The ask

Bring one Cortex use case and your consumption commitment.

Ten working days. Which workload to build first, what the per-call premium adds to the credit bill, and what owning the same workflow costs to build. You keep the analysis.

Start an ignite buildSee how we build
Fixed feeTen working daysNo commitment beyond the diagnostic