HR Chatbot / Employee Self-Service

Part of the worlds-biggest-software-project initiative.

An open-source, AI-native HR assistant that answers policy questions, retrieves payslips and leave balances, and executes employee self-service transactions against live HRIS data.

This project is a self-hostable conversational HR platform built around a pre-built HR intent library and Model Context Protocol (MCP) tool calls into HRIS systems such as Workday, BambooHR, SAP SuccessFactors, and ADP. It targets HR Operations, IT, and engineering teams at organisations that are too small for Moveworks-class enterprise contracts but need more HR depth than a generic chatbot framework provides.

Why HR Chatbot / Employee Self-Service?

Moveworks (acquired by ServiceNow in March 2025 for $2.85B) starts at roughly $200K–$1M+/year and requires 5,000+ employees, leaving the entire SMB and mid-market segment unserved.
Leena AI and Workativ are proprietary SaaS with per-employee or session-based pricing, and integration depth is shallower than the enterprise tier.
Generic OSS frameworks like Rasa (Apache 2.0) and BotPress (MIT) ship no pre-built HR intents — every leave-balance query, payslip lookup, and policy answer must be built from scratch.
Static FAQ-style bots routinely give stale or generic policy answers; an AI-native tool can query the real HRIS in real time and cite exact figures.
GDPR, CCPA, and SOC 2 expectations push regulated buyers toward self-hosted deployments where conversation data never leaves the organisation's infrastructure.

Key Features

Pre-Built HR Intent Library

Leave balance queries and leave-request submission
Payslip retrieval and pay-related questions
Personal detail updates (address, emergency contact, banking)
Org chart lookups and manager / reportee resolution
Policy Q&A grounded in uploaded handbooks and HR documents

Live HRIS Integration via MCP

Model Context Protocol tool calls for structured, auditable HRIS access
Connectors for Workday, BambooHR, ADP, and SAP SuccessFactors
Real-time data retrieval rather than static exports or stale FAQ caches
RAG over uploaded HR policy documents and employee handbooks with cited passages
Custom action endpoints for organisation-specific workflows

Multi-Channel Employee Access

Slack and Microsoft Teams native deployment
WhatsApp and SMS channels for deskless and frontline workers
Web widget for embedding in HRIS portals and intranets
Multi-language conversational support (English, Spanish, French, German, Portuguese in v1.1)
Graceful escalation to a human agent with full conversation context

Enterprise Security and Compliance

SAML 2.0 and OIDC SSO for identity verification before personal-data access
Self-hosted deployment so all conversation data stays inside the organisation's infrastructure
GDPR-aligned data minimisation and right-to-erasure design
WCAG 2.1 AA accessibility for regulated industry and government buyers
Aligned with SOC 2 Type II and ISO/IEC 27001 operational expectations

Analytics and Proactive Engagement

Query volume, resolution rate, and deflection metrics dashboard
Top unresolved-query categorisation for HR ops teams
Proactive notifications for open enrollment, performance review, and expiring leave balances
Employee satisfaction tracking on conversational interactions
Unhandled-intent reporting to drive iterative improvement

AI-Native Advantage

Retrieval-Augmented Generation over live HRIS data means the bot can answer "how much leave do I have left?" with the actual figure pulled from Workday in real time, not a stale cached value. Personalised policy answers are tailored to the employee's employment type, location, seniority, and benefit elections rather than returning generic handbook text. AI-driven proactive nudges replace email blasts with contextual prompts triggered by HR system events, and an MCP-first architecture makes every HRIS action structured and auditable rather than ad hoc.

Tech Stack & Deployment

The project is designed for self-hosted deployment so conversation data never leaves the organisation's infrastructure, with optional managed-cloud configurations for teams that prefer hosted operation. Integration is built around Model Context Protocol for structured HRIS tool calls, with REST connectors for Workday, BambooHR, ADP, and SAP SuccessFactors. SSO uses SAML 2.0 and OIDC. The LLM layer is provider-agnostic: OpenAI, Anthropic Claude, Azure OpenAI, or local models via Ollama can serve as the language backbone.

Market Context

The global HR chatbot / conversational AI market was valued at approximately $2.1B in 2024 and is projected to reach $6.5B by 2030, a roughly 20% CAGR (research.md). Incumbent pricing ranges from ~$299/mo for Workativ Starter and $5–$15 PEPM for mid-market SaaS up to $200K–$1M+/year for Moveworks and ServiceNow HR. Primary buyers are VPs and Directors of HR Operations focused on ticket deflection, CIOs and CISOs evaluating data residency and SSO, and CHROs tracking employee experience metrics.

Project Status

This project is in the research and specification phase.
Contributions, feedback, and domain expertise are welcome.

Contributing

We welcome contributions from developers, domain experts, and potential users. See CONTRIBUTING.md for guidelines.

Important: All contributions must be your own original work or clearly attributed open-source material with a compatible licence. Copyright infringement and licence violations will not be tolerated and will result in immediate removal of the offending contribution. If you are unsure whether a piece of code, text, or other material is safe to contribute, open an issue and ask before submitting.

Licence

Licence to be determined. See discussion for context.