Project Overview
What is Yappa Knowledge Hub?
Yappa Knowledge Hub is an AI-powered internal knowledge management tool designed for the team at Yappa. It solves the problem of scattered information across Slack messages, emails, notes, and deal updates by providing a centralised system for capturing, processing, and distributing knowledge.
Internship Context
| Detail | Value |
|---|---|
| Institution | UCLL University of Applied Sciences |
| Programme | Professional Bachelor |
| Duration | 56 working days (~12 weeks, 3 months) |
| Company | Yappa |
| Project | Internal Knowledge Hub (Kennisverzameltool) |
| Current Status | Week 3 - Sprint 1 MVP in progress |
POC Complete
Sprint 0 POC completed with 24 work items resolved (pending merge). Sprint 1 now focuses on AI features with demo target: end-to-end Slack → AI → Slack flow.
Problem Statement
Internal knowledge at Yappa is currently scattered across Slack messages, forwarded emails, notes, deal updates, announcements, and external content. Information shared "in the moment" becomes hard to:
- Find later — no central search across knowledge sources
- Contextualise — different teams need different perspectives on the same information
- Convert into updates — weekly overviews and digests require manual effort
- Keep curated — no moderation or quality control mechanism
Vision
Build a central system that makes internal knowledge:
- Easy to capture — low-friction submission directly from Slack
- Smartly processed — AI summaries tailored per audience role (developer, CEO, marketer)
- Actively distributed — periodic digests by thematic list and target group
Core Concepts
Thematic List
A curated stream of resources designed for a specific purpose and audience. Each list has a name, description, target audience roles, default tags, and a digest schedule.
Resource
A single knowledge item submitted by a user — text, URL, PDF, or audio transcript. Each resource carries metadata, tags, and can be assigned to multiple lists.
Target Group
A role-based audience (e.g., Developers, Marketers, CEO) that determines how AI summaries are styled through per-group prompt templates.
Digest
A periodic bundled report per list, generated on schedule (weekly/biweekly/monthly), containing all new resources with their summaries, delivered via Slack.
Scope — 13 Epics, 100 User Stories
The project is structured across 13 epics totalling 100 user stories and 302 story points:
| # | Epic | Stories | SP |
|---|---|---|---|
| 01 | Content Ingestion | 12 | 39 |
| 02 | Thematic Lists & Organisation | 9 | 22 |
| 03 | AI-Powered Summaries | 10 | 29 |
| 04 | Periodic Digests / Reports | 6 | 23 |
| 05 | Bot / Service Interaction | 8 | 24 |
| 06 | Configuration & Administration | 8 | 19 |
| 07 | Search & Discovery | 7 | 23 |
| 08 | User / Role Management | 7 | 18 |
| 09 | Analytics & Insights | 7 | 18 |
| 10 | Content Quality & Moderation | 6 | 12 |
| 11 | Integrations & Extensibility | 6 | 24 |
| 12 | Onboarding & Help | 5 | 11 |
| 13 | AI Agent & Orchestration | 9 | 40 |
| Total | 100 | 302 |
Priority Breakdown
- 🔴 Must-have (MVP) — 32 stories (core Slack ingestion, AI summaries, digests, config)
- 🟡 Should-have — 47 stories (search, analytics, moderation, extensibility)
- 🟢 Nice-to-have — 21 stories (advanced agents, exports, bulk import)
Tech Stack
| Layer | Technology | Purpose |
|---|---|---|
| Backend | PHP 8.x + Symfony 7 | API, workers, command bus |
| Database | PostgreSQL / MySQL | Relational data (Doctrine ORM) |
| Queue | Redis + Symfony Messenger | Async processing |
| AI | OpenRouter / OpenAI / local | Provider-agnostic summarisation |
| Platform | Slack (Google Workspace) | Primary user interface |
| Docs | VitePress + Mermaid | This documentation site |
Stakeholders
Success Metrics
| Metric | Target | Measurement Method |
|---|---|---|
| Active users (weekly) | 20+ | Unique Slack user IDs submitting knowledge |
| Knowledge items | 100+ | Count in database |
| Digest open rate | 60%+ | Slack message metrics |
Audience Roles (Target Groups)
Each resource can be tailored for specific audiences, which determines the AI's summarization style:
- Developers: Focus on implementation, API details, and technical debt.
- Marketers: Focus on market impact, messaging, and feature benefits.
- CEO/Leadership: Focus on strategic alignment, ROI, and high-level progress.
- Service Desk: Focus on support steps, known issues, and user impact.
- Sales: Focus on selling points, competitive edge, and pricing.
Technical Infrastructure
The MVP architecture is designed for scalability and resilience:
- Redis: Persistent state for view tracking and session management.
- Symfony Messenger: Asynchronous job processing for AI and scraping.
- Monolog: Structured logging for debugging and audit trails.
- Health Checks: Automated monitoring of system components.
Roadmap
Strategic plan for development:
- Sprint 0 POC (Complete): Foundation architecture and core features.
- Sprint 1 Infrastructure: Production-ready environment and monitoring.
- Sprint 2-5 Features: AI enhancements, digests, and advanced search.
- Sprint 6 Polish: Bug fixes, performance tuning, and launch.
For a detailed timeline, see the MVP Roadmap.
Stakeholders
| Decision | Choice | Rationale |
|---|---|---|
| Primary channel | Slack | Where the team already works |
| AI strategy | Provider-agnostic | Flexibility to switch models |
| Ingestion UX | Message Shortcut + Modal | Lowest friction, no commands to memorise |
| Processing | Async (Messenger) | Non-blocking Slack interactions |
| MVP scope | Pasted text + AI summaries + Digests | Highest value, lowest complexity |