Skip to main content
SnowflakePatent grant

Snowflake was granted a patent for verifying LLM-generated SQL queries

Source: Autobound Signal API · 148 signals tracked for Snowflake · updated

Get Snowflake's live signals in Claude or ChatGPT, or by API

AskWhat changed at Snowflake this week? Use Autobound signals for snowflake.com.

What happened

In November 2026 Snowflake patented cloud data-platform technology for checking AI-generated queries before they retrieve information from connected databases.

Source

patents.google.comNov 17, 2026

US patent US12801599

LLM-generated text-to-SQL verification via directional graphs

Abstract

Described is a system for updating an LLM-generated SQL query preventing double counting by receiving a natural language query from a user at a cloud-based server, where the query requests access to data in a database. The data platform identifies semantic data from the query and, using a large language model (LLM), generates a SQL query containing join functions. The SQL query is then parsed to identify operation types and data sources. An undirected graph is constructed from the parsed query, with nodes representing data sources and edges representing join functions. The undirected graph is converted to a directed graph by adding relationship characteristics to the edges. Based on this directed graph, the data platform validates the SQL query to detect and resolve potential double counting issues and updates the SQL query as necessary to ensure accurate results.

patents.google.com/patent/US12801599Read the full source

Extracted by Autobound

From the Signal API record
Signal
Patent grant

What this signalsA new patent often shows where a company is putting its R&D budget.

Grant date
Nov 17, 2026
Inventors
Yahia Khaled El Bsat
Tech area
software / computing
Patent number
US12801599

The full record

From the Signal API record

People

  • Yahia Khaled El BsatInventor

Details

Primary CPC class
G06F 16/24522 (Electric digital data processing)
USPTO assignee
Snowflake Inc.
What the invention does
A method that turns an AI-written SQL query into a graph of its tables and joins to spot paths that could count the same data twice.

Topics and mentions

Tags

  • Intellectual Property
  • Research & Development
  • Artificial Intelligence
  • Cloud
  • Database

Technologies named

  • text-to-SQL
  • large language models
  • graph analysis
  • cloud databases

Extraction

Detected
Nov 18, 2026
signal_type
patents-company
signal_subtype
patentGrant

Use this data

Get every patent signal for Snowflake and the companies you sell to, in the tools you already use.

  1. Ask Claude about it

    Connect Autobound to Claude, Claude Code or Cursor with MCP. Then ask: “What changed at Snowflake this week?”

  2. Send it to your own tools

    The Signal API returns patent signals for any list of companies as JSON, for your CRM, warehouse or app.

  3. Try it free

    Sign up and spend your free credits on the companies you sell to.

    Start Free1,000 free credits

The API returns more than this page shows

This page shows a preview. The full patents-company record in the Signal API and MCP can also have these 8 fields. Some fields are empty for some signals.

Company

  • linkedin_urlValue in the API
  • industriesValue in the API
  • employee_count_lowValue in the API
  • employee_count_highValue in the API
  • revenueValue in the API
  • descriptionValue in the API

Signal

  • signal_nameValue in the API
  • associationValue in the API
Show the full JSONThe record on this page and the API request

GET /v1/signals/9f186a88-a606-5c87-b355-fb2897943255 returns this record as JSON. POST /v1/companies/enrich returns every signal for snowflake.com.

{
  "signal_id": "9f186a88-a606-5c87-b355-fb2897943255",
  "signal_type": "patents-company",
  "signal_subtype": "patentGrant",
  "detected_at": "2026-11-18T15:30:52+00:00",
  "company": {
    "name": "Snowflake",
    "domain": "snowflake.com"
  },
  "data": {
    "cpc": [
      {
        "code": "G06F 16/24522",
        "label": "Electric digital data processing"
      },
      {
        "code": "G06F 16/243",
        "label": "Electric digital data processing"
      },
      {
        "code": "G06F 16/2456",
        "label": "Electric digital data processing"
      },
      {
        "code": "G06F 16/258",
        "label": "Electric digital data processing"
      },
      {
        "code": "G06F 16/9024",
        "label": "Electric digital data processing"
      }
    ],
    "tags": [
      "Intellectual Property",
      "Research & Development",
      "Artificial Intelligence",
      "Cloud",
      "Database"
    ],
    "detail": "In November 2026 Snowflake patented cloud data-platform technology for checking AI-generated queries before they retrieve information from connected databases.",
    "summary": "Snowflake was granted a patent for verifying LLM-generated SQL queries",
    "event_at": "2026-11-17",
    "inventors": [
      {
        "name": "Yahia Khaled El Bsat",
        "status": "resolved",
        "is_primary": true,
        "linkedin_url": "https://www.linkedin.com/in/yahia-bsat"
      }
    ],
    "tech_area": "software / computing",
    "event_kind": "grant_date",
    "grant_date": "2026-11-17",
    "source_url": "https://patents.google.com/patent/US12801599",
    "cpc_primary": {
      "code": "G06F 16/24522",
      "label": "Electric digital data processing"
    },
    "patent_title": "LLM-generated text-to-SQL verification via directional graphs",
    "patent_number": "US12801599",
    "uspto_assignee": "Snowflake Inc.",
    "patent_abstract": "Described is a system for updating an LLM-generated SQL query preventing double counting by receiving a natural language query from a user at a cloud-based server, where the query requests access to data in a database. The data platform identifies semantic data from the query and, using a large language model (LLM), generates a SQL query containing join functions. The SQL query is then parsed to identify operation types and data sources. An undirected graph is constructed from the parsed query, with nodes representing data sources and edges representing join functions. The undirected graph is converted to a directed graph by adding relationship characteristics to the edges. Based on this directed graph, the data platform validates the SQL query to detect and resolve potential double counting issues and updates the SQL query as necessary to ensure accurate results.",
    "precision_class": "structural",
    "invention_explanation": "A method that turns an AI-written SQL query into a graph of its tables and joins to spot paths that could count the same data twice.",
    "technologies_mentioned": [
      "text-to-SQL",
      "large language models",
      "graph analysis",
      "cloud databases"
    ]
  }
}

Long text fields are shortened on this page.

Looking up one signal by its id is free. Enrich costs 2 credits per signal returned; a call with no results is free.