Landbase
Toolkit · Advanced Dataset Generator
← All tools
T8 · Tool

Advanced Dataset Generator

Express any go-to-market question as Spark SQL across the Landbase graph and any joined external datasets. Semantic search, AI qualification, and enrichment are callable functions inside a query.

Examples

How customers use it

01
Combining title and job-history conditions
Ask for "companies with 3+ current SDRs whose past job history includes Outreach or Salesloft AND who currently use Salesforce + Gong." Filter UIs return "no results" or silently drop conditions; SQL answers the question exactly as asked.
02
Asking negative and absence questions
Want "B2B SaaS, $20M+ revenue, with marketing-ops headcount but no demand-gen leader." Most platforms can't express the negative; SQL checks presence and absence on the same row and returns only the matching companies.
03
Joining customer data at query time
Bring your own data — a partner list, an intent feed, a fund-portfolio CSV. The Dataset Generator joins it against the Landbase graph at query time so the unified table can be scored, filtered, and qualified.
04
Running AI qualification as a column
Need the full qualification pass on 5,000 accounts, including a clause like "must be a freight forwarder per their website" that requires per-record judgment. The Dataset Generator embeds AI Qualification directly into the query so the column populates with Yes/No/Unknown plus evidence as it runs.
Pairs with
T1Semantic Search T5Enrich T6AI Qualification T4Match
Found in
Complex Queries TAM Mapping
Under the hood+
How it's used

Nebulous, multi-clause queries. Title-array transforms across full job history. Department-absence predicates. Joins against external datasets the customer brings to the table. The kind of question other vendors return "no results" for, or quietly drop the hard part of.

Why it matters

If a customer can phrase the question, the engine can answer it. No filter-UI ceiling. No "we can't slice it that way." Reproducible, auditable, version-controlled. The other seven tools are callable from inside a query — which is what makes SQL on this stack more than just SQL.

How it works
  • Spark SQL — executed against the 200M-entity graph and any joined external datasets.
  • Custom functions — free-text matching, semantic transforms, custom aggregations all available as callable building blocks.
  • Tools-as-callable — Semantic Search, AI Qualification, and Enrich are invokable from inside a query.
  • Reproducibility — every query is preserved with its result snapshot.