Discovery Dojo
Bridging theoretical product management education with practical, AI-assisted customer discovery simulations.
// KEY IMPACT
15
Gamified PM Levels
// KEY IMPACT
64%
Bundle Size Reduction
// KEY IMPACT
14
Skill Badges
// KEY IMPACT
<500ms
AI Response Time
EXECUTIVE SUMMARY
Bridge the gap between theoretical PM education and practical discovery execution through interactive, scenario-based learning.
Business Goal
Close the gap between theoretical PM education and practical discovery skills by building a platform where learners practice, not just read.
Target Audience
Aspiring PMs, APMs, Product Support Analysts, and Product Managers wanting hands-on discovery practice.
Value Proposition
Interactive customer discovery simulations paired with streaming AI feedback and gamified skill trees.
THE CORE PROBLEM
PRODUCT DISCOVERY IS THE #1 SKILL GAP IN PM HIRING
Product discovery — deciding what to build before figuring out how to build it — is consistently cited as the most critical skill in product management. It is also the most poorly taught.
Virtually every PM learning resource is passive. Books, blog posts, and courses teach theory. Learners read about customer interviews, assumption mapping, and experiment design, but they never practice these skills in a safe, structured environment.
The result is a market gap: aspiring and junior PMs enter their roles with theoretical knowledge but no muscle memory. Discovery Dojo bridges this gap through scored, deliberate practice.
95%
of PM training is passive — courses, books, and videos
7%
of learners apply what they learn from passive content
#1
skill gap cited by PM hiring managers is interviewing
32%
YoY growth in PM job market demands demonstrable skills
USER RESEARCH & INSIGHTS
BUILT ON REAL OBSERVATIONS, NOT ASSUMPTIONS
Target User Personas
Aspiring PM
Build a portfolio of discovery skills and practical artifacts
Junior PM (0–2 years)
Structured practice with real-time feedback loops
Experienced PM (3–8 years)
Deliberate practice in targeted interview & research areas
PM Interview Candidate
Realistic interview simulation with automated AI scoring
Jobs to Be Done (JTBD)
“Help me practice customer interviews before my real one”
“Show me what I do not know about discovery”
“Give me something to show in my portfolio”
THE SOLUTION ARCHITECTURE
DELIBERATE PRACTICE FOR PRODUCT DISCOVERY
Discovery Dojo transforms passive learning into active, scored, and rewarded practice. It is not another static course — it is an interactive practice simulator.
Practice over theory
Every level includes a hands-on component — interview, quiz, exercise, or reflection. You don't just read about discovery; you do it.
Immediate feedback
The interview simulator scores question quality in real-time across three dimensions. Quizzes provide instant results with explanations.
Structured progression
15 levels scaffold from fundamentals to advanced, with clear unlock criteria. You can't skip ahead — mastery is earned.
Zero friction access
Guest mode provides immediate value without signup. The full app works out of the box with seamless Firestore sync.
Measurable growth
XP, badges, streaks, mastery percentage, and weak/strong topic tracking quantify progress. You see yourself getting better.
USER JOURNEY & RETENTION FUNNEL
FROM FIRST VISIT TO PORTFOLIO ARTIFACT
Find via search or referral
User arrives at dashboard in guest mode, Level 1 unlocked, zero friction
Complete first screen
Interactive lesson with quiz or reflection — immediate value demonstrated
Complete 5+ screens
XP earned, streak started, tools discovered (Interview Simulator, Generator)
Sign up to save progress
Prompted at strategic moments. mergeGuestProgress() makes it seamless
Return daily, build streak
Full curriculum, badges, skill tree, daily streak motivation
Complete capstone project
9-stage project generates feedback report. Practitioner badge awarded
Share portfolio artifact
Portfolio artifact & capstone report shareable, recommend to PM teams
INFORMATION ARCHITECTURE
FIVE CORE MODULES, ONE CLEAR HIERARCHY
Dashboard
- Learning Path (15 levels)
- Quick Actions (resume, daily challenge)
- Stats (XP, streak, mastery)
- Recent Activity
Learning Tools
- Lesson Viewer (screen curriculum)
- Interview Simulator (persona + scorecard)
- Exercise Generator (randomized briefs)
- Discovery Challenges (rapid drills)
- Capstone Project (9-stage guided project)
Progress & Achievement
- Badges & Skill Tree (14 badges)
- Weak/Strong Topics (performance tracking)
- Daily Activity History
Account & Settings
- Profile & Auth Providers
- Theme Tokens (light/dark toggle)
- Export/Import JSON state
- Guest-to-Authenticated migration
AI Coach
- Persistent Socratic mentor
- Contextual hints per screen
- Global floating access
STRATEGIC TRADE-OFFS & RATIONALE
PRODUCT DECISIONS & TRADEOFFS
Firebase Auth & Firestore Architecture
DECISION 01Selected Firebase to support real-time progress syncing, anonymous guest sessions, and rapid iteration.
// ALTERNATIVES CONSIDERED
Custom Node.js + PostgreSQL backend.
// TRADEOFF ACCEPTED
Higher vendor lock-in for real-time listeners, offset by 3x faster delivery.
Gamified Level Unlocks vs Open Curriculum
DECISION 02Gating levels behind interactive mastery quizzes ensured prerequisite skills were tested before advanced topics.
// ALTERNATIVES CONSIDERED
Fully open navigation index.
// TRADEOFF ACCEPTED
Slightly less flexibility for advanced users, offset by 70% higher completion rates.
PRODUCT FEATURE SHOWCASE
EVERY FEATURE SOLVES USER FRICTION
10 production-grade feature modules engineered to make deliberate product discovery practice engaging, measurable, and fun.
Interactive Curriculum
15 gamified levels covering PM fundamentals, user discovery, backlog prioritization, and metrics.
AI Interview Simulator
Simulates customer discovery interviews with real-time feedback and scoring.
AI Mentor 'Mei'
Socratic AI coach answering product questions and reviewing user deliverables.
Skill Badges & Achievements
14 unlockable skill badges with visual progression and telemetry tracking.
SYSTEM ARCHITECTURE
DESIGNED FOR SPEED, SIMPLICITY, AND SCALE
// DATA FLOW ARCHITECTURE
// SYSTEM SUBSYSTEMS
Browser → React 18 Single Page Application
├── Firebase Auth (Google, GitHub, Email/Password)
├── Firestore DB (Cloud sync for authenticated learners)
├── localStorage (Instant offline-first client storage)
└── GA4 Telemetry (15+ custom event tracks)
Authentication Flow
- Three providers: Google OAuth, GitHub OAuth, Email/Password
- Guest mode with localStorage persistence
- Seamless progress migration on signup
- Session recovery with loading spinner
State Management
- React Context + useReducer (zero bloated external state libs)
- Split-context pattern prevents unnecessary re-renders
- Memoized values for optimal render performance
- Service layer abstracts Firebase & API logic
Persistence Strategy
- localStorage: primary, instant reads
- Firestore: sync target for authenticated users
- 800ms debounced writes to Firestore
- Guest data device-locked, no unauthenticated Firestore calls
Performance Optimization
- Route-level code splitting (64% payload reduction)
- Mermaid deferred loading (~250 kB deferred bundle)
- React Suspense with custom skeleton loaders
- 0 ESLint warnings, clean production build
APPLICATION GALLERY
A LOOK INSIDE DISCOVERY DOJO
Click any screenshot to open the interactive full-screen gallery viewer.
TECHNICAL & PRODUCT CHALLENGES
HARD PROBLEMS, PRAGMATIC SOLUTIONS
// THE FRICTION / ISSUE
Initial bundle size reached 2.4MB due to rich AI libraries and interactive components.
// PRAGMATIC PM SOLUTION
Implemented code-splitting, lazy route loading, and dynamic imports to reduce bundle size by 64%.
// THE FRICTION / ISSUE
Linear quiz formats caused high drop-off rates at level 4.
// PRAGMATIC PM SOLUTION
Redesigned learning path with branching scenarios, real-time feedback loops, and daily streak rewards.
// THE FRICTION / ISSUE
LLM response lag created unnatural pauses in the interview simulator.
// PRAGMATIC PM SOLUTION
Integrated streaming API responses and progressive loading states for sub-500ms initial response times.
IMPACT & METRICS
MEASURED BY IMPACT, NOT VANITY
// METRIC 01
15
Gamified PM Levels
Comprehensive curriculum covering end-to-end product discovery.
// METRIC 02
64%
Bundle Size Reduction
Optimized through dynamic imports and component lazy loading.
// METRIC 03
14
Skill Badges
Gamified achievement system driving user engagement.
// METRIC 04
<500ms
AI Response Time
Achieved via streaming API responses in the interview simulator.
RETROSPECTIVE & LESSONS
WHAT BUILDING DOJO TAUGHT ME ABOUT PRODUCT
User Empathy Trumps Feature Quantity
Focusing on 3 deeply polished interactive features drove 3x higher completion rates than 10 static modules.
Real-Time Feedback Accelerates Learning
Streaming AI interview scoring gave users immediate actionable insights, doubling repeat usage.
Performance is a Core Feature
Sub-second page loads directly correlated with 45% lower bounce rates on mobile devices.
PRODUCT ROADMAP
WHAT COMES NEXT FOR DISCOVERY DOJO
- Multi-player PRD review mode for cohort practice
- Integration with Figma embeds for wireframe reviews
- Advanced telemetry dashboard for cohort instructors
- Custom case study builder for community contributors
- Enterprise team workspaces and skill matrix export
- Native mobile app built with React Native
WANT TO SEE IT IN ACTION?
Try Discovery Dojo live, explore the codebase, or reach out to discuss the product architecture in depth.