FoundryOps API
Entity Resolution at Scale
Bring FoundryOne™ intelligence directly into your data pipelines, warehouses, and applications. Real-time matching, batch deduplication, and master-data unification — powered by the same engine trusted by RevOps and data-ops teams.
Built for Engineers, Not Dashboards
One intelligence layer for all your entity-resolution needs — from lead routing to customer 360 to analytics deduplication.
Core Capabilities
Real-Time & Batch Matching
Handle sub-100ms single-record lookups or multi-million-row batch jobs with the same precision and explainability.
Programmatic Matching & Dedupe
REST and gRPC endpoints for fuzzy search, record linkage, and deduplication. Plug directly into Airflow, dbt, or your CI/CD data pipelines.
Customer Master Data Store
A high-performance key–value and graph index that maintains your "golden record" of companies and contacts. Serve identity resolution across all downstream systems.
Full / Partial / Delta Updates
Ingest complete datasets, update individual entities, or stream incremental changes. Optimized for warehouses (Snowflake, BigQuery, Databricks) and event streams (Kafka, Pub/Sub).
Complete API Toolkit
SDKs in Python, Node.js, and Go · OpenAPI 3 spec · Postman collections · sandbox environment · reference notebooks.
Graph-Powered Intelligence
Access Foundry Graph™ — 2M+ companies, 27k+ LEI-verified, 50k+ parent/subsidiary relationships, and multilingual aliases. Query or embed it to resolve corporate families programmatically.
One API, Three Match Modes
From sub-second lookups to million-row batch jobs — choose the mode that fits your workload.
Transactional
/api/v2/match/record
Real-time lookup for lead routing and inline product UX. Sub-2s latency target.
- 1 record, ~128KB payload
- Immediate scores + reason chips
- Idempotent with Idempotency-Key
Batch
/api/v2/match/batch
Up to 100 records per call. Async job with poll endpoint or webhook callback.
- 100 records, 512KB max
- callback_url for push delivery
- Progress polling + ETA
File-Scale
/api/v2/match/files
CSV or Parquet via presigned upload. No hard row cap — built for warehouse-scale.
- Presigned upload → storage_key
- Results as JSONL or Parquet
- Concurrency + queue guardrails
Managed Reference Data
Upload, version, and activate your master datasets with schema enforcement, data quality summaries, and snapshot management.
Schema Enforcement
Define typed schemas (int, string, timestamp). Auto-coercion for common types. Rejected rows written to NDJSON.gz artifacts.
Data Quality Reports
Accept/reject counts, null percentages, type coercion stats, PK coverage. Know your data before you match against it.
Append & Snapshot
Append inline JSON or uploaded files. Each append creates a new snapshot. Activate any snapshot for matching.
Audit & RBAC
Full audit trail for create/append/activate/delete. Role-based access controls on datasets, jobs, and webhooks.
Tunable Match Profiles
Pre-configured profiles for common use cases — or dial in your own thresholds.
Minimize false positives. Best for automated routing.
Default. Good for most deduplication workflows.
Cast a wide net. Review candidates manually.
Optimized for CRM L2A matching.
Transparent Billing
Shadow Mode
Default for development. Usage is metered but no credits are debited. See simulated costs via headers.
X-Credits-Remaining-SimulatedEnforced Mode
Production billing. Credits debited per operation. 402 response with top-up link when balance is low.
X-Credits-RemainingWant the full specs and code?
Explore detailed endpoints, curl examples, dataset v2 flows, billing, and SDK snippets in the developer docs.
Explore the DocsIntegrate Everywhere
From pipelines to no-code automations — FoundryOps API plugs into your favorite platforms.
Zapier / Make.com
Connect Salesforce, HubSpot, or Sheets to FoundryOps for instant lead-to-account matching. No-code RevOps automations.
Pipedream / n8n
Run entity-resolution workflows as serverless functions. Visual builders with code when you need it.
Airflow / dbt / Snowflake
Schedule batch dedupe or enrichment inside your existing data pipelines. Drop FoundryOps into any DAG.
Whether you write Python or build Zaps, FoundryOps meets you where you work.
Why FoundryOps API
Most teams eventually build entity resolution themselves — and maintain it forever. Others stitch together vendor-specific APIs that never agree on what "the same company" means. FoundryOps API gives you one deterministic, explainable layer across all use cases.
- Single engine, many workloads — lead routing · analytics dedupe · customer 360 · data activation
- Built for B2B complexity — multi-brand, multinational, alias-rich datasets
- Transparent, not black box — every match includes reason chips and confidence metrics
- Privacy-safe by design — open-data sourcing (Wikidata, Companies House, SEC EDGAR), SOC 2 ready, no brokered data
Common Use Cases
CRM Lead-to-Account Matching
Resolve inbound leads to existing accounts in real time; trigger routing or scoring via webhook.
Analytics Warehouse Deduplication
Batch-dedupe dimensions inside Snowflake, BigQuery, or Databricks for reliable reporting.
Customer 360 / MDM
Build a unified customer record across CRM, billing, support, and product-usage systems.
Embedded Matching in Your Product
Expose FoundryOne-powered company/contact matching inside your SaaS app with a single API call.
Use FoundryOne Today
While the API is in early access, you can start using the same intelligence layer through:
FoundryOps for Google Sheets — instant matching and enrichment in spreadsheets
FoundryOps Web (1H 2026) — browser-based RevOps Data Hygiene suite
Stop Building Entity Resolution In-House
Join the early-access list and be the first to integrate FoundryOne™ intelligence into your data stack.
Launching 2H 2026 • Be first in line