BambooHR AI: Native Features, API and HR Automations
Understand BambooHR native AI and its API, then design HR review apps, mapping agents, and approved workflows on Major for onboarding and PTO reconciliation.

BambooHR AI includes Bamboo AI for workforce insights and actions, plus Ask BambooHR for employee questions. This guide explains those features, the BambooHR API, and where Major adds governed, cross-system automation.
For HR operations and HRIS leaders, Major provides deterministic apps that hold review queues and history, agents that judge ambiguous cases, and workflows that coordinate approved actions across systems.
What BambooHR does for HR operations
BambooHR holds employee records, reporting relationships, time off, hiring information, and other people data. HR administrators maintain records and reporting, managers review team information, and employees use self-service within their access levels. Its API introduction describes reading and modifying employee data and generating reports.
The operational difficulty often sits between systems. Recruiting may finish in Greenhouse while employee administration begins in BambooHR. Payroll operations may need to compare approved leave with UKG Pro before a pay run. A completed action in one product does not prove the receiving team has the right information.
Major is the enterprise platform where agents build the software they run on. For these handoffs, the software is an operations app with a managed database, approval views, and action receipts. An agent interprets unfamiliar department labels or explains a reconciliation exception. Code handles known mappings, validation, and record matching, so established administrative rules do not need fresh reasoning on every run.
BambooHR AI and its built-in capabilities
Bamboo AI is BambooHR's native intelligence layer. The vendor describes natural-language workforce questions, charts and custom reports, follow-up analysis, recommended next steps, and routine HR actions. Check feature availability in your account rather than assuming every marketed capability is enabled.
Ask BambooHR focuses on employee self-service. It answers questions about policies, benefits, employee information, and organizational structure from HRIS data and documents. The product page describes source citations, role-based access, desktop and mobile access, a Slack experience, and downloadable Q&A reporting.
BambooHR also documents beta AI connectors and an MCP foundation, controlled by an administrator-enabled AI Connectors app. Those vendor capabilities do not establish a native Major connector.
Use the native features for questions and supported actions within BambooHR. Major adds a durable control surface when work spans recruiting, payroll, documents, and approvals. Its agent can assess conflicting evidence across those systems, while the app preserves the case and the workflow enforces the authorized next step. Employment decisions remain with accountable people.
How BambooHR connects to Major
BambooHR has no native or Pipedream connector in the catalog: it connects as a custom API connector, with base URL and authentication configured once by IT. Major orchestrates BambooHR; it never replaces it.
The current API base is https://{companyDomain}.bamboohr.com/api/. Configure the company subdomain explicitly. Do not mix employee identifiers across tenants: BambooHR documents employee IDs as immutable and unique within a company.
IT establishes the connector once, stores credentials centrally in Major's connector vault, and restricts access to the required operations. Credentials belong in neither code nor prompts. SSO, RBAC, and audit logs cover the apps, agents, and workflows using that connection.
For a first implementation, separate read-only reconciliation from employee creation. The queue app needs the evidence and approval state, not unrestricted access to every HR field. Give the mapping agent only the source values and permitted target choices. Let the workflow execute the approved payload through the connector, using the stored case identifier to connect the write with its authorization.
BambooHR API guide for apps, agents, and workflows
Authentication, pagination, events, and testing
For a single-company integration, API-key authentication uses HTTP Basic Auth: the key is the username and an arbitrary string is the password. Access follows the associated enabled user's permissions. OAuth applications use bearer access tokens; a refresh token requires the offline_access scope.
Endpoint-specific OAuth scopes include employee reads and writes, field metadata, change tracking, and time off. Request only the scopes needed for the selected operations, including relevant field-level scopes, and test the actual user's access. A scope does not justify assuming visibility into every employee.
List Employees pagination uses page[after], page[before], and page[limit]. The default limit is 250, with a maximum of 2,500. Save the returned cursor or follow the response's next link. Other endpoints have their own filters, not a universal pagination scheme.
A numerical global request quota is not documented publicly in the technical overview. Handle HTTP 429 and honor Retry-After. X-BambooHR-Error-Message can help diagnose failures, but its text should not drive program logic.
Event-based webhooks notify employee creation, monitored-field updates, and deletion. Updated events require monitor fields. Verify the SHA-256 HMAC signature over the raw body followed by X-BambooHR-Timestamp, using the private key, and compare against X-BambooHR-Signature. Network failures and server errors trigger up to five retries: immediately, then after 5, 10, 20, and 40 minutes, with jitter.
The getting-started guide explicitly calls for a test BambooHR account. Use synthetic employees to exercise permissions, duplicates, missing fields, and retry behavior before production.
GET /api/v1/meta/fields: discover the schema
Returns standard and custom employee field definitions, including identifiers, names, types, and deprecation information. The reference lists employee and field OAuth scopes.
- App: Store the reviewed field mapping and schema version used by each case.
- Agent: Interpret an unfamiliar source label against the permitted target definitions.
- Workflow: Refresh metadata before validating newly introduced mappings.
GET /api/v1/employees: retrieve the roster
Returns a cursor-paginated employee list. The reference lists employee and narrower field-related OAuth scopes; restricted fields can be null and marked in the response.
- App: Maintain permitted employee references for reconciliation screens.
- Agent: Explain a possible identity mismatch without treating unreadable data as absent.
- Workflow: Traverse pages and checkpoint a roster refresh after persistence.
GET /api/v1/employees/{id}: fetch one employee
Returns a known employee's requested fields. With no fields parameter, only the ID is returned; inaccessible fields can be omitted silently. The reference documents employee and field-specific scopes.
- App: Show the authorized current values beside a proposed correction.
- Agent: Compare current job information with an approved source document.
- Workflow: Re-read relevant fields immediately before an approved update.
POST /api/v1/employees: create a record
Creates an employee, with first and last name as the documented minimum. The reference lists employee.write and field-specific write scopes; payroll-linked configurations require additional fields.
- App: Retain the approved payload, source application ID, and resulting employee ID.
- Agent: Propose mappings for ambiguous role or department information before approval.
- Workflow: Create only after authorization, then store the response before downstream tasks.
POST /api/v1/employees/{id}: update selected fields
Updates employee field values using POST, not PATCH. The reference lists employee.write and field-specific write scopes. Employment type and termination date require employment-status table handling instead.
- App: Present the precise before-and-after field comparison for review.
- Agent: Explain why a proposed correction conflicts with existing information.
- Workflow: Submit the approved fields and verify the resulting record.
GET /api/v1/employees/changed: catch up changes
Returns changed employee IDs since a timestamp, with change type and last-change time. The reference lists public.integration and public.user OAuth scopes.
- App: Hold the sync watermark and each employee's last successful refresh.
- Agent: Triage cases where changes conflict with an open review.
- Workflow: Poll using the since parameter and advance the watermark only after durable processing.
GET /api/v1/time_off/requests: read leave requests
Returns requests overlapping start and end dates, updated since a timestamp, or matching either condition when combined. Its documented OAuth scope is time_off.
- App: Keep the pay-period request snapshot and reconciliation outcome.
- Agent: Explain differences in dates, request status, units, or category mapping.
- Workflow: Fetch the payroll window and process relevant approved requests.
GET /api/v1/employees/{employeeId}/time_off/calculator: check balances
Returns balances across visible categories as of the end date, under the time_off OAuth scope. Time Off tab permissions apply independently from request visibility.
- App: Display the dated balance evidence used in a payroll exception.
- Agent: Distinguish an accrual discrepancy from a discretionary policy's normal zero balance.
- Workflow: Retrieve each permitted employee's balance for the agreed comparison date.
BambooHR automation use cases on Major
These are proposed designs, not claims of preinstalled integrations. Each additional system needs its own authorized connection and verified operations.
Greenhouse and BambooHR onboarding handoff
- App: A staging queue holds the source application, mapped fields, approval, and employee ID.
- Agent: Map unfamiliar department and location labels, flagging missing information rather than guessing.
- Workflow: Detect a hired application, validate it, obtain HR approval, then create the employee.
- BambooHR endpoints: List Fields, List Employees, and Create Employee.
BambooHR and UKG Pro PTO reconciliation
- App: A payroll-period ledger stores employee mappings, request snapshots, and unresolved differences.
- Agent: Assess whether a difference reflects timing, category interpretation, or missing evidence.
- Workflow: Read both systems before payroll cutoff, route exceptions to payroll operations, and record the resolution without automatically changing balances.
- BambooHR endpoints: List Time Off Requests and Get Time Off Balance.
BambooHR, Google Drive, and Slack record corrections
- App: A restricted correction queue stores a source-document reference and proposed field changes.
- Agent: Compare an authorized job-change document with the employee's current record and explain conflicts.
- Workflow: Refresh changed employees, request HR review through a private notification, then submit approved corrections. Slack receives a case link, not sensitive document contents.
- BambooHR endpoints: Get Changed Employee IDs, Get Employee, and Update Employee.
Build the BambooHR architecture on Major

Figure 1. Separate connector access, orchestration, judgment, and durable state.
Build the operations app first: a managed database for cases, source references, approved mappings, action attempts, and receipts. Its UI is the review queue and the record of what happened. The app never decides whether an ambiguous employee match is correct.
The workflow accepts a verified notification or schedule tick, fetches scoped context, validates known rules, and calls the agent only when interpretation is needed. The agent proposes a mapping or exception classification with supporting evidence. A routing step then sends the case to the appropriate reviewer.
- Configure the custom BambooHR connector and the other authorized system connections.
- Build the app's case records, review views, and operation ledger.
- Give the agent a narrow judgment task and a permitted output structure.
- Add workflow triggers, validation, retries, and approval gates.
- Test the full path with synthetic records and restricted users.
Store approved reusable mappings in the app. Later matching cases can use deterministic code rather than asking the agent to reinterpret the same labels. That is the practical meaning of reason once, run forever.
A worked example with illustrative numbers
For example, suppose HR processes 100 onboarding handoffs per month. Manual preparation takes 12 minutes per hire: locating the offer, checking fields, re-entering values, and recording completion. That is 100 × 12 = 1,200 minutes, or 20 hours of preparation.
Assume a Major implementation reduces human preparation and review to 3 minutes for each of 80 straightforward records. The remaining 20 exceptions take 8 minutes each because HR must resolve ambiguous mappings or missing evidence. The new workload is 80 × 3 + 20 × 8 = 400 minutes, or 6 hours and 40 minutes.
The difference is 800 minutes, or 13 hours and 20 minutes monthly. These are illustrative assumptions, not customer results or a promise. They exclude implementation, maintenance, and elapsed waiting time.
For each hire, the app stores the case, the agent handles only unfamiliar mappings, and the workflow waits for HR approval before creation. Measure actual review time and exception frequency before using this calculation in a business case. A faster average is not useful if duplicate records or unsupported payroll fields require cleanup later.
Governance, approvals, and failure handling

Figure 2. Reads can refresh automatically; consequential writes require authorized review.
Define read-only paths separately from writes. The agent must never independently approve employment decisions, change compensation, terminate employment, override leave balances, or disclose sensitive personnel information. For administrative corrections, approval should bind to the exact payload and source version. Recheck material source changes before execution.
Persist notifications before acknowledging them and deduplicate using company, employee, event timestamp, and payload identity. This is an implementation design, not a vendor-provided exactly-once guarantee. Use changed-ID polling as a reconciliation path for missed notifications.
For permissioned webhooks, user deactivation stops delivery and permission changes can remove fields. Alert on stale syncs. Never reinterpret access loss as an employee deletion.
On 429, honor the documented retry header; use bounded backoff for transient failures. Route invalid fields and permission errors to an owner. After a timed-out create, reconcile before resubmitting, since the original write may have succeeded. Keep attempt history, reviewer identity, payload version, and response receipts in the app's audit record, with retention and redaction appropriate to HR data.
BambooHR AI Apps, Agents and Workflow Automations
Every published Major use case that works with BambooHR, newest first. Each one names the BambooHR API objects involved and shows the app, agent and workflow split.
- Greenhouse BambooHR Integration: Automate New-Hire Onboarding: Stop re-keying accepted offers. Design a controlled Greenhouse-to-BambooHR handoff that prepares employee records, catches mismatches, and gives HR a clear approval point before day one.
Frequently asked questions
What is BambooHR AI?
BambooHR AI includes Bamboo AI for workforce insights and routine actions, plus Ask BambooHR for employee questions grounded in HR data and policies. Availability varies by feature and account. Major adds governed cross-system processes with durable app state, agent judgment, and workflow approvals.
Does BambooHR have an AI assistant?
Yes. Ask BambooHR answers questions about policies, benefits, employee information, and organizational structure. BambooHR describes citations, role-based permissions, desktop and mobile access, and a Slack experience. Confirm which information your organization has made available before relying on an answer.
How does BambooHR connect to Major?
BambooHR connects through a custom API connector, not a native or Pipedream catalog connector. IT configures the company-specific base URL and authentication once. Major keeps credentials centrally scoped, while the app, agent, and workflow receive only the access needed for their process.
Can a BambooHR AI agent create employee records?
A Major agent can propose field mappings and identify missing evidence. In the design here, the workflow creates the employee only after an authorized HR reviewer approves the payload. The app retains the source identifier, approval, employee ID, and action receipt to prevent duplicate handoffs.
Can BambooHR automation reconcile PTO with payroll?
Yes, a proposed Major workflow can compare permitted BambooHR requests and dated balances with payroll data from an authorized UKG Pro connection. The agent explains discrepancies; payroll operations approves any correction. An empty request result is not proof of no leave because permissions can limit visibility.
Does BambooHR offer API testing and webhooks?
BambooHR's getting-started guide calls for a test account. Its documentation also describes employee-change webhooks, HMAC verification, and retries for network failures and server errors. Test signatures, permission loss, duplicate delivery, and recovery from missed notifications before enabling production writes.
Sources
The following official pages were reviewed for this guide. API reference pages provide the method, path, parameters, and OAuth scope declarations for the operations above. The product pages establish the native AI capabilities, while the technical guides explain authentication, delivery behavior, and testing. Recheck them when configuring your tenant because documentation and feature availability can change.
- BambooHR documentation index and machine-readable page directory
- Getting started: authentication, OAuth, API keys, and test accounts
- Technical overview: company base URL, errors, and throttling
- List Fields: standard and custom employee field discovery
- List Employees: roster pagination and restricted field behavior
- Get Employee: requested fields and employee access rules
- Create Employee: required values and payroll configuration caveats
- Update Employee: POST updates and employment-status limitations
- Get Changed Employee IDs: timestamp-based incremental change tracking
- List Time Off Requests: date windows and visibility caveats
- Get Time Off Balance: dated calculations and policy permissions
- Webhooks: signature verification, request format, and retry schedule
- Permissioned webhooks: field access and user deactivation behavior
- Event-based webhooks: employee creation, updates, and deletion
- Field-based webhooks: payload contents and event behavior
- BambooHR and AI: beta connectors and administrator access controls
- Ask BambooHR: employee self-service and cited HR answers
- Bamboo AI: native workforce intelligence and routine actions
Start building on Major
Build a BambooHR review queue, a field-mapping agent, and an approval workflow that keeps HR records authoritative and cross-system work visible.
About the author
Ritvij Gautam is Head of Growth at Major.
Last reviewed November 20, 2026 by Ritvij Gautam.
Frequently asked questions
- What is BambooHR AI?
- BambooHR AI includes Bamboo AI for workforce insights and routine actions, plus Ask BambooHR for employee questions grounded in HR data and policies. Availability varies by feature and account. Major adds governed cross-system processes with durable app state, agent judgment, and workflow approvals.
- Does BambooHR have an AI assistant?
- Yes. Ask BambooHR answers questions about policies, benefits, employee information, and organizational structure. BambooHR describes citations, role-based permissions, desktop and mobile access, and a Slack experience. Confirm which information your organization has made available before relying on an answer.
- How does BambooHR connect to Major?
- BambooHR connects through a custom API connector, not a native or Pipedream catalog connector. IT configures the company-specific base URL and authentication once. Major keeps credentials centrally scoped, while the app, agent, and workflow receive only the access needed for their process.
- Can a BambooHR AI agent create employee records?
- A Major agent can propose field mappings and identify missing evidence. In the design here, the workflow creates the employee only after an authorized HR reviewer approves the payload. The app retains the source identifier, approval, employee ID, and action receipt to prevent duplicate handoffs.
- Can BambooHR automation reconcile PTO with payroll?
- Yes, a proposed Major workflow can compare permitted BambooHR requests and dated balances with payroll data from an authorized UKG Pro connection. The agent explains discrepancies; payroll operations approves any correction. An empty request result is not proof of no leave because permissions can limit visibility.
- Does BambooHR offer API testing and webhooks?
- BambooHR's getting-started guide calls for a test account. Its documentation also describes employee-change webhooks, HMAC verification, and retries for network failures and server errors. Test signatures, permission loss, duplicate delivery, and recovery from missed notifications before enabling production writes.