BambooHR AI-Powered Company Policies & Benefits Chatbot

Unlock Instant HR Answers with this AI Agent

This workflow transforms your HR support by providing an intelligent chatbot that directly integrates with BambooHR and your company's policy documents. Employees can ask questions in natural language and receive accurate, context-aware answers regarding company policies, benefits, and even find contact details for specific colleagues or department heads. This AI Agent leverages OpenAI's language models for understanding queries, retrieving relevant information from documents stored in a Supabase vector database, and interacting with BambooHR for real-time employee data.

Key Features & Benefits

• Automated Knowledge Base Creation: Automatically fetches policy documents (PDFs) from a designated BambooHR folder, processes them (text splitting, embedding with OpenAI), and stores them in a Supabase vector store for efficient retrieval.
• Natural Language HR Queries: Utilizes OpenAI LLMs to understand and respond to a wide range of employee questions about handbook content, 401k policies, benefits, expense guidelines, and more.
• Integrated Employee Lookup: Features an AI-driven tool that queries BambooHR to find employee details (job title, department, supervisor) by name or identify senior department members.
• Contextual Conversations: Maintains conversation history (Window Buffer Memory) for more natural and helpful interactions.
• Retrieval Augmented Generation (RAG): Combines the power of LLMs with information retrieved from your specific company documents, ensuring answers are relevant and accurate.
• AI-Driven Automation: Streamlines HR support, reduces manual query handling, and empowers employees with self-service access to information 24/7.

• OpenAI API Key with access to suitable models (e.g., gpt-3.5-turbo, gpt-4 for chat and text-embedding-ada-002 or similar for embeddings).
• BambooHR API Key and your company subdomain.
• Supabase project credentials (URL and Anon Key) for vector storage. You'll need to create a table and a PL/pgSQL function for matching documents (e.g., match_files).
• Company policy documents (ideally PDFs) stored in BambooHR under a consistently named category (e.g., 'Company Files').

1. Download the workflow JSON file.
2. Configure BambooHR Integration (Initial Document Load):
   • In the 'GET all files' (BambooHR) node, set up your BambooHR API credentials and company subdomain.
   • Adjust the 'Filter out files from undesired categories' node: In its conditions, ensure the rightValue matches the exact name of the BambooHR category where your policy documents are stored (e.g., ='Company Files').
3. Configure OpenAI Credentials:
   • In all 'Embeddings OpenAI' and 'OpenAI Chat Model' nodes, select or create your OpenAI API credentials.
4. Configure Supabase Vector Store:
   • In the 'Supabase Vector Store' (insert mode) and 'Supabase Vector Store Retrieval' nodes, configure your Supabase API credentials (URL & Anon Key).
   • Set the tableName parameter to your desired table name (e.g., company_files).
   • Ensure the queryName parameter (e.g., match_files) corresponds to a vector similarity search function you've created in your Supabase database.
5. Review AI Agent Configuration:
   • Examine the 'HR AI Agent' node: Customize the systemMessage to refine the AI's persona, objectives, and guidelines for tool usage to match your company's specifics.
6. Understand Employee Lookup Tool:
   • The 'Employee Lookup Tool' (toolWorkflow node) calls back into this same workflow, triggering the 'AI-Powered HR Benefits and Company Policies Chatbot' (ExecuteWorkflowTrigger) node to use its dedicated logic path for employee/department lookups. Ensure the BambooHR nodes within this path ('GET all employees', etc.) are correctly configured with credentials.
7. Initial Document Processing:
   • Manually run the workflow using the 'When clicking ‘Test workflow’' trigger. This will fetch, process, and embed your company policies into the Supabase vector store. Monitor for any errors.
8. Activate Chatbot:
   • Once document processing is successful, activate the main workflow and the 'Employee initiates a conversation' (Chat Trigger) node to make the chatbot live.