.png)
Eightfold AI Integration Guide
Integrate Eightfold AI with HR, recruiting, and workforce platforms through tenant-specific APIs, callbacks, exports, and orchestrated data workflows.
Eightfold AI integration options at a glance
Eightfold AI integrations are typically based on tenant-specific REST APIs, although endpoint coverage, authentication, and supported operations must be confirmed with the customer’s implementation documentation. Selected tenants or modules may also provide webhook-style callbacks, bulk or asynchronous exchange, scheduled exports, or files for larger synchronization workloads. Martini can consume documented Eightfold APIs, receive supported callbacks, process vendor-provided exports, transform Candidates, Employees, Jobs, Requisitions, Applications, and Skills, and expose normalized APIs for downstream applications. Credentials and environment-specific settings can be stored securely while workflows manage validation, checkpoints, retries, reconciliation, and auditability.
| Integration point | Supported by Eightfold AI? | Common use cases | How Martini supports it |
|---|---|---|---|
| REST APIs | Limited | Eightfold AI is commonly integrated through enterprise APIs for Candidates, Employees, Jobs, Requisitions, Applications, and Skills, but endpoint coverage and operations are tenant- or contract-dependent. | Martini can consume documented Eightfold REST endpoints, generate reusable integration assets from supplied API definitions, map payloads, apply rules, and expose normalized APIs. |
| Webhooks and outbound callbacks | Limited | Selected Eightfold products or tenants may provide callbacks for candidate, employee, job, application, or other events; a general event catalog was not confirmed. | Martini can receive documented callback requests through a workflow trigger, validate and deduplicate them, transform the payload, and route it to downstream systems. |
| Bulk, asynchronous, and batch exchange | Limited | Large populations may be exchanged through tenant-specific bulk services, asynchronous export jobs, scheduled files, pagination, or incremental filters. | Martini can schedule bounded batch workflows, persist checkpoints, poll documented job status endpoints, and reconcile partial results. |
| File and attachment exchange | Not confirmed | Resumes, candidate documents, job descriptions, or other recruiting content may be exchanged through APIs or scheduled files, but a public attachment specification was not verified. | Martini can process supported files or route binary and encoded content when the customer’s Eightfold implementation documents the delivery mechanism. |
| Authentication | Limited | Authentication varies by API product and tenant and may involve service accounts, API credentials, OAuth 2.0, scopes, authorization headers, or network controls. | Martini stores confirmed credentials and environment-specific settings as secrets and applies the authentication scheme specified by Eightfold. |
| GraphQL APIs | Not confirmed | No official Eightfold GraphQL API documentation was verified, so GraphQL should not be assumed for the integration. | Martini can consume GraphQL generally, but an Eightfold GraphQL integration should only be implemented if the customer supplies confirmed documentation. |
| SOAP APIs | Not confirmed | No official Eightfold SOAP API documentation was verified for the researched integration scope. | Martini supports SOAP generally, but Eightfold SOAP endpoints should not be used unless confirmed by the tenant documentation. |
| Database and analytics access | Not confirmed | Direct production database access was not verified. Analytics or workforce-planning data may instead be available through product-specific reports or exports. | Martini can consume supported reports, exports, APIs, or files, but should not connect directly to Eightfold production databases without explicit vendor support. |
How Eightfold AI exposes data and business events
Eightfold AI REST APIs
Eightfold AI is commonly integrated with HR, recruiting, and workforce platforms through APIs. The exact resources, operations, base URL, authentication, and permissions must be confirmed from the customer’s tenant documentation or API agreement.
Martini implementation pattern
Martini implementation pattern: Martini consumes the documented Eightfold REST API, retrieves or submits supported objects, maps fields into canonical models, applies validation and business rules, and records identifiers, checkpoints, and outcomes for reconciliation.
Implementation sequence
Eightfold AI callbacks
Eightfold AI may provide webhook-style callbacks or outbound notifications for selected products, modules, or events. A general event catalog and universal coverage across Candidates, Employees, Jobs, Applications, and Requisitions were not confirmed.
Martini implementation pattern
Martini implementation pattern: Martini exposes a receiving API or workflow trigger, validates the callback according to the confirmed tenant contract, checks event identifiers for duplicate delivery, and routes the normalized event to the required application or queue.
Implementation sequence
Eightfold AI batch and exports
Large Employee, Candidate, Job, Application, or Skills populations may require pagination, bulk services, asynchronous exports, scheduled files, or tenant-specific reporting. The available mechanism must be confirmed for the deployment.
Martini implementation pattern
Martini implementation pattern: A scheduled workflow processes bounded pages or files, persists a checkpoint or change marker, transforms records, and separates transient failures from validation errors for controlled retry and reconciliation.
Implementation sequence
Common Eightfold AI integration patterns
Pattern 1: Synchronize employees from an HRIS
When to use this pattern
Use this pattern when Workday, SAP SuccessFactors, Oracle HCM Cloud, UKG Pro, or another HR system is the source of Employees and organizational data for Eightfold AI. It supports scheduled incremental exchange while keeping ownership of worker master data explicit.
Integration direction
Example Mapping
| Eightfold AI Field | Canonical Field | Target Field |
|---|---|---|
| workerId | employee.externalId | Eightfold Employee external identifier |
| jobProfile | employee.jobTitle | Eightfold Employee job or role |
| organization | employee.organizationId | Eightfold Employee organization |
| skills | employee.skills | Eightfold Skills |
Martini implementation pattern
A scheduler starts the workflow, which retrieves changed workers, validates required identifiers and controlled taxonomies, maps the payload to confirmed Eightfold operations, and submits changes. Martini maintains a checkpoint and synchronization ledger, applies field ownership rules, and sends authorization, validation, or ambiguous outcomes to reconciliation instead of blindly retrying them.
Martini capabilities used
- workflows
- scheduling
- API consumption
- data mapping
- business rules
- secrets management
- error handling
Pattern 2: Synchronize candidates and applications with an ATS
When to use this pattern
Use this pattern when Eightfold AI and Greenhouse or another ATS exchange Candidates, Applications, Jobs, and recruiting activity. It is appropriate for bidirectional synchronization only after ownership, stable identifiers, and supported operations are agreed.
Integration direction
Example Mapping
| Eightfold AI Field | Canonical Field | Target Field |
|---|---|---|
| candidateId | candidate.externalId | Greenhouse candidate ID |
| applicationStatus | application.status | Greenhouse application status |
| jobId | application.jobExternalId | Greenhouse job ID |
| skills | candidate.skills | Greenhouse candidate skills |
Martini implementation pattern
Martini retrieves or receives supported candidate and application data, correlates related Candidates and Jobs, filters sensitive fields, and maps records into the ATS model. Before creates, the workflow looks up stable identifiers or ledger entries; transient failures use bounded backoff while uncertain outcomes are reconciled before another create attempt.
Martini capabilities used
- API consumption
- workflow orchestration
- data mapping
- business rules
- duplicate detection
- retry handling
- audit logging
Pattern 3: Publish approved jobs and requisitions
When to use this pattern
Use this pattern when approved Requisitions or Jobs originate in an HR or recruiting platform and must be submitted or updated in Eightfold AI. It supports controlled publication and reconciliation of rejected or unpublished items.
Integration direction
Example Mapping
| Eightfold AI Field | Canonical Field | Target Field |
|---|---|---|
| requisitionId | requisition.externalId | Eightfold Requisition identifier |
| title | job.title | Eightfold Job title |
| location | job.location | Eightfold Job location |
| employmentType | job.employmentType | Eightfold Job employment type |
Martini implementation pattern
A workflow retrieves approved items, normalizes locations and employment types, maps Skills and job requirements, and applies publication eligibility rules. Martini stores returned Eightfold identifiers, records rejected payloads with error details, and retries only transient failures.
Martini capabilities used
- workflows
- API consumption
- mapping and transformation
- validation
- business rules
- error handling
- reconciliation
Pattern 4: Process Eightfold callback events
When to use this pattern
Use this pattern when the Eightfold tenant provides documented callbacks for selected candidate, employee, job, or application events. It is useful for lower-latency updates, but event availability, ordering, authentication, and delivery semantics must be confirmed first.
Integration direction
Example Mapping
| Eightfold AI Field | Canonical Field | Target Field |
|---|---|---|
| eventId | event.id | ServiceNow correlation ID |
| eventType | event.type | ServiceNow workflow trigger |
| objectId | talentObject.externalId | ServiceNow record reference |
| updatedAt | event.occurredAt | ServiceNow event timestamp |
Martini implementation pattern
Martini receives the callback, validates its authentication or signature, checks the event ledger, and retrieves the current object when the notification lacks sufficient data. The workflow transforms the event into a normalized ServiceNow request, queues or processes it, returns the required acknowledgment, and routes downstream failures for retry or manual reconciliation.
Martini capabilities used
- API exposure
- workflow triggers
- data validation
- data mapping
- deduplication
- queue or asynchronous execution
- error handling
Applications commonly integrated with Eightfold AI
Eightfold AI commonly participates in enterprise HR, recruiting, and workforce architectures. The exact objects, operations, and direction of exchange depend on the customer’s Eightfold modules, APIs, permissions, and agreements.
| Application | Scenario | Direction | Martini Pattern |
|---|---|---|---|
| Workday | Synchronize Employees, organizational data, Jobs, and recruiting information between the HRIS and Eightfold AI. | Workday → Martini → Eightfold AI | A scheduled Martini workflow retrieves changed worker and organization data, validates required fields, maps it to the documented Eightfold API model, submits supported changes, and records identifiers and outcomes for reconciliation. |
| SAP SuccessFactors | Exchange employee, job, requisition, and recruiting data for talent intelligence and workforce processes. | SAP SuccessFactors → Martini → Eightfold AI | Martini can orchestrate incremental HR data retrieval, taxonomy normalization, validation, and tenant-specific API submission, with retry and exception handling for rejected records. |
| Oracle HCM Cloud | Provide worker, job, organization, and recruiting data to Eightfold AI and return supported talent or matching results. | Oracle HCM Cloud → Martini → Eightfold AI | A workflow maps Oracle HCM Cloud payloads into Eightfold structures, applies ownership and eligibility rules, and maintains a synchronization ledger for replay and audit. |
| Greenhouse | Synchronize Jobs, Candidates, Applications, and recruiting workflow data between the recruiting platform and Eightfold AI. | Greenhouse → Martini → Eightfold AI | Martini can coordinate bidirectional exchanges using stable external identifiers, transform recruiting taxonomies, prevent duplicate creates, and route validation failures for reconciliation. |
| Support recruiting workflows involving professional profiles, sourcing, or job distribution where applicable permissions and products allow the exchange. | LinkedIn → Martini → Eightfold AI | Where permitted interfaces are available, Martini can mediate approved profile or job data, filter sensitive fields, apply consent and ownership rules, and pass only supported data to Eightfold AI. | |
| ServiceNow | Connect employee-service, workforce, or internal mobility processes with Eightfold AI talent intelligence. | Eightfold AI → Martini → ServiceNow | Martini can normalize Eightfold employee or mobility data, apply routing rules, and expose or consume APIs for ServiceNow workflows with correlation identifiers and controlled retries. |
| UKG Pro | Synchronize worker, organization, job, and workforce data with Eightfold AI. | UKG Pro → Martini → Eightfold AI | A scheduled workflow retrieves bounded batches from UKG Pro, maps worker and organizational taxonomies, submits confirmed Eightfold operations, and stores checkpoints for incremental processing. |
| ADP Workforce Now | Exchange worker and employment data to support talent and workforce processes. | ADP Workforce Now → Martini → Eightfold AI | Martini can transform approved ADP Workforce Now data into the tenant’s Eightfold contract, enforce field-level filtering, and send failures to a reconciliation path rather than retrying invalid payloads. |
How to build a Eightfold AI integration in Martini
Objective
Confirm the Eightfold tenant contract before building the integration, including base URL, API products, permissions, credentials, scopes, network controls, and sandbox availability.
Instructions in Martini
- Obtain the tenant-specific API specification or implementation guide
- Confirm whether the integration uses service credentials, API credentials, OAuth 2.0, or another documented method
- Store confirmed credentials and environment settings as Martini secrets
Objective
Select a schedule, API request, callback, or file-based trigger based on the confirmed Eightfold capability and required synchronization latency.
Instructions in Martini
- Use a scheduler for incremental or batch synchronization
- Use a workflow trigger for documented Eightfold callbacks
- Define the checkpoint, event identifier, or file naming convention
Objective
Retrieve or receive Eightfold objects using documented resources, pagination, exports, or callbacks while preserving enough metadata for replay and reconciliation.
Instructions in Martini
- Process bounded pages or batches
- Persist cursors, timestamps, job status, or file checkpoints
- Capture request identifiers and source object identifiers
Objective
Coordinate related objects such as Candidates, Jobs, Applications, Requisitions, Employees, and Skills in a Martini workflow rather than embedding the logic in point-to-point scripts.
Instructions in Martini
- Resolve related identifiers before writing dependent objects
- Separate transient failures from validation and authorization failures
- Use conditional routing for create, update, skip, and reconciliation paths
Objective
Transform Eightfold payloads and enterprise source data into canonical and target-specific models while accommodating tenant-specific field names and taxonomies.
Instructions in Martini
- Map stable identifiers and controlled values
- Normalize locations, employment types, organizations, and Skills
- Filter sensitive candidate and employee fields according to the use case
Objective
Enforce ownership, eligibility, duplicate-prevention, privacy, and publication rules before submitting changes or forwarding events.
Instructions in Martini
- Use external IDs, lookup checks, or idempotency keys where supported
- Validate required fields and tenant-specific enumerations
- Prevent blind retries of uncertain create operations
Common Eightfold AI data objects used in integrations
| Object | Typical Use | Common target systems | Martini handling |
|---|---|---|---|
| Candidates | External applicants and prospective talent profiles used in recruiting, sourcing, and talent intelligence workflows. | Greenhouse, Workday, SAP SuccessFactors, Oracle HCM Cloud, LinkedIn | Martini maps confirmed candidate fields, applies privacy and duplicate-prevention rules, and synchronizes using stable external identifiers where the API supports the operations. |
| Employees | Current workers used for internal mobility, skills analysis, and workforce processes. | Workday, SAP SuccessFactors, Oracle HCM Cloud, UKG Pro, ServiceNow | Martini processes incremental employee changes, filters sensitive fields, validates organizational values, and records checkpoints and target identifiers. |
| Jobs | Job postings and opportunities presented to candidates or distributed through recruiting processes. | Greenhouse, Workday, SAP SuccessFactors, LinkedIn | Martini normalizes locations, employment types, requirements, and Skills before submitting or updating documented Eightfold operations. |
| Requisitions | Hiring requests and associated recruitment requirements used to initiate or manage hiring workflows. | Workday, SAP SuccessFactors, Oracle HCM Cloud, Greenhouse | Martini maps requisition ownership and status, applies publication rules, and routes rejected or unpublished items to reconciliation. |
| Applications | Candidate submissions or application activity associated with Jobs and recruiting workflows. | Greenhouse, Workday, SAP SuccessFactors, Oracle HCM Cloud | Martini correlates Candidates, Jobs, and Applications, uses idempotency controls where available, and avoids blind retries of create operations. |
| Skills | Skills, competencies, and inferred capabilities used in matching, talent intelligence, and workforce planning. | Workday, SAP SuccessFactors, Oracle HCM Cloud, UKG Pro, ServiceNow | Martini maintains configurable taxonomy mappings, validates controlled values, and applies tenant-specific rules before synchronization. |
Authentication and security considerations
Tenant-specific authentication
Eightfold AI authentication is not universal across the researched integration scope. Confirm whether the tenant uses service accounts, API credentials, OAuth 2.0, authorization headers, scopes, IP allowlisting, or other controls.
Secure implementation
- Store confirmed credentials and environment-specific configuration in Martini secrets.
- Use least-privilege permissions and separate development, test, and production environments.
- Protect candidate and employee data with TLS-protected communication, field filtering, minimal retention, and appropriate log redaction.
- Capture correlation identifiers without exposing sensitive payload content unnecessarily.
Operational considerations for Eightfold AI integrations
Tenant and schema variation
API resources, operations, taxonomies, rate limits, and event availability can vary by Eightfold module, tenant, geography, and contract. Confirm the API specification, sandbox, permissions, and supported objects before implementation.
Volume and synchronization
- Confirm page limits, cursors, continuation tokens, export jobs, and incremental change markers.
- Persist checkpoints and process Candidates, Employees, Jobs, Applications, and Skills in bounded batches.
- Use stable identifiers, idempotency controls where supported, and a synchronization ledger to prevent duplicate writes.
- Apply bounded retries and backoff for throttling or transient failures; do not blindly retry uncertain create operations.
Events, files, and testing
- Confirm callback authentication, event identifiers, ordering, acknowledgment, retry, and replay behavior.
- Verify whether resumes and documents use binary content, encoded data, temporary URLs, or scheduled files.
- Test controlled taxonomy changes, schema changes, rejected records, and partial batch failures before production deployment.
Why use Martini instead of scripts or point-to-point integrations?
Orchestration instead of isolated scripts
Martini provides a maintainable workflow layer around Eightfold AI APIs, callbacks, exports, and files. It separates authentication, retrieval, transformation, business rules, target writes, and operational handling so the integration can evolve with tenant-specific contracts.
Reusable integration assets
- Consume documented Eightfold REST APIs and reuse mappings and workflow services across HR, recruiting, and workforce processes.
- Expose normalized APIs so downstream applications do not each implement Eightfold-specific contracts.
- Support scheduled, event-driven, batch, and API-led patterns in one integration environment.
- Centralize validation, retries, checkpoints, reconciliation, monitoring, and environment-specific secrets.
Frequently asked questions
Eightfold AI can be integrated through tenant-specific REST APIs and, where enabled, webhook-style callbacks, bulk or asynchronous exchange, scheduled exports, or files. The exact objects, operations, authentication, and event coverage must be confirmed for the customer’s modules and agreement.
Yes. Martini can consume documented Eightfold REST APIs, receive supported callback events, process vendor-provided exports or files, map Eightfold objects to enterprise models, and expose normalized APIs. A native Martini Eightfold AI connector was not verified.
No. A dedicated Eightfold AI connector is not required. Martini can use Eightfold’s confirmed native APIs, callbacks, authentication methods, exports, or files through workflows, APIs, mappings, transformations, and business rules.
Lonti does not charge an additional per-connector or per-vendor fee to integrate Eightfold AI. The integration is subject to the provisioned capacity of the Martini environment. Separate costs may apply from Eightfold AI, infrastructure, or other third-party systems based on subscription, usage, and deployment model.
Use documented tenant-specific REST APIs as the primary option when available. Consider callbacks for selected events and bulk, asynchronous, scheduled, or file-based exchange for larger volumes. GraphQL, SOAP, direct database access, and a universal public API should not be assumed.
Some Eightfold products or tenants may provide webhook-style callbacks, but a general event catalog and universal coverage were not confirmed. Confirm supported events, authentication or signing, retry behavior, acknowledgment requirements, ordering, and replay semantics before relying on event-driven processing.
Martini can run scheduled or event-driven workflows that retrieve or receive Eightfold Candidates, Employees, Jobs, Requisitions, Applications, and Skills, then map them to canonical and target models. Checkpoints, stable identifiers, controlled taxonomies, and synchronization ledgers support incremental processing and reconciliation.
Martini can distinguish transient failures from validation and authorization errors, apply bounded retries and backoff, capture request identifiers, and route unresolved items to reconciliation. Stable external IDs, lookup-before-create logic, idempotency keys where supported, and duplicate-event detection help prevent duplicate Candidates, Jobs, Applications, or Requisitions. Martini can also expose a normalized REST API for downstream applications.
Related Martini documentation
APIs
Workflows
Transformation
Connect Eightfold AI with your enterprise systems
Use Martini to implement secure, maintainable Eightfold AI integrations around confirmed APIs, callbacks, exports, and files.