Bluewoo HRMS - Build Guide

This documentation describes what we will build - a modern, AI-native HR Management System for small to medium businesses.

What We Will Build

Bluewoo HRMS will be an AI-powered HRMS SaaS platform combining:

• Core HR management features
• AI-native assistant with RAG capabilities
• Multi-tenant architecture with complete data isolation

Target Outcomes

When complete, the platform will achieve:

Core Features to Build

Employee Management

• Users will be able to create, edit, and view employee profiles
• System will support organizational hierarchies (manager-employee relationships)
• Basic org chart visualization

Time-Off Management

• Users will be able to submit leave requests
• Managers will be able to approve/reject requests
• System will track balances automatically

Document Management

• Users will be able to upload and manage HR documents
• System will support AI-powered document search via RAG
• Access control by role and department

Team Communication

• Instagram-style feed for team updates
• Users will be able to post text, images, video, voice
• AI will generate post summaries

Goals / OKR (Basic)

• Users will be able to set personal, team, and company goals
• AI will suggest goals based on role/description
• Progress tracking (0-100%)

Workflows

• Onboarding/offboarding checklists
• Task tracking and assignment

Analytics Dashboard

• Headcount metrics
• Team activity summaries
• AI-generated insights

AI Assistant

• Natural language queries about HR data
• Policy Q&A via RAG
• Document search across all files

Technology Decisions

Constraints: Prisma 5.x (NOT 6.x), Express 4.x (NOT 5.x), Redis 7.x (NOT 8.x), Zod 3.x

Architecture Decision

Browser (Next.js 15) ◄──► HRMS App (NestJS 10) ◄──► AI Service (Express)
                              │                          │
                         PostgreSQL 17              MongoDB Atlas
                              │
                           Redis 7

• Monolithic HRMS + separate AI service
• REST APIs only (no GraphQL)
• PostgreSQL for HRMS data, MongoDB for AI vectors only
• Multi-tenant with row-level isolation

Documentation Structure

Goals & Use Cases (Chapters 1-10)

• Product Vision - What we will build and why
• Tech Stack - Technology decisions
• Domain Model - Core entities
• Database - Data storage goals
• API Design - API goals
• Backend Architecture - Backend structure
• Frontend Architecture - Frontend structure
• Multi-Tenancy - Isolation goals
• Authentication - Auth goals
• AI Service - AI capabilities

Rules & Guidelines (Chapters 11-14)

• Development Standards - How to code
• Testing - Quality gates
• Deployment - How to deploy
• Extension Guide - How to extend

Implementation (Chapters 15-17)

• Implementation Plan - Detailed building blocks
• AI Coding Agents - Patterns, constraints, workflow
• UI Screenshots - UI reference

How to Use This Documentation

For Humans

1. Read Chapters 1-10 to understand WHAT we will build
2. Read Chapters 11-14 for rules and guidelines
3. Use Chapter 15 for step-by-step execution

For AI Agents

Start with AI Coding Agents:

• Patterns - Code patterns to follow
• Constraints - Rules to obey
• Workflow - How to work

Success Criteria

The project will be complete when:

• All building blocks in Chapter 15 pass their gates
• Users can manage employees, time-off, documents
• AI assistant answers HR queries via RAG
• Multi-tenant isolation is verified
• System handles target load (500+ tenants)