Meeting Intelligence Platform - Solution

This is the solution repository for the meeting intelligence interview problem.

Purpose

This repository contains:

• ✅ Fully implemented models.py
• ✅ Complete meeting_service.py with all phases implemented
• ✅ All tests passing
• 📋 INTERVIEWER_GUIDE.md with interview instructions

Quick Start

1. Install Dependencies

uv sync

2. Run Tests

# Test individual phases
uv run python run_tests.py phase1
uv run python run_tests.py phase2

# Test all phases
uv run python run_tests.py all

All tests should pass ✅

For Interviewers

See INTERVIEWER_GUIDE.md for:

• How to conduct the interview
• What to look for in each phase
• Discussion prompts and follow-up questions
• Time management guidelines
• Evaluation criteria

Project Structure

meeting_ci_question_solution/
├── README.md                 # This file
├── INTERVIEWER_GUIDE.md      # Interview guide for you
├── problem.md                # Same as candidate sees
├── pyproject.toml            # Python dependencies
├── data.py                   # Mock database
├── llm_client.py            # Mock LLM client
├── models.py                # ✅ Complete models
├── meeting_service.py       # ✅ Complete implementation
├── run_tests.py             # Test runner
└── tests/                   # Test files
    ├── test_phase1.py
    ├── test_phase2.py
    ├── test_phase3.py
    ├── test_phase4.py
    ├── test_phase5.py
    └── test_all.py

What's Implemented

Phase 1: Pydantic Models

• Complete User, Contact, Meeting models
• All optional fields added
• Proper type hints

Phase 2: CRUD Operations

• get_all_meetings() with filtering
• get_meeting() with error handling
• create_meeting() with validation
• update_meeting() with partial updates
• delete_meeting() with validation

Phase 3: Availability Algorithm

• Time interval merging
• Gap detection
• Work hours filtering
• Multiple user support
• Edge case handling

Phase 4: Pre-Meeting Prep

• Historical meeting fetching
• LLM prompt construction
• JSON response parsing
• Structured output generation

Phase 5: Error Handling

• Exponential backoff retry logic
• LLM failure handling
• Proper exception management
• Graceful degradation

Using This During Interview

1. Before interview: Review INTERVIEWER_GUIDE.md
2. During interview: Have this repo open for reference
3. Compare: Check candidate's approach vs solution
4. Discuss: Use as basis for architectural discussions

Verification

All tests pass:

uv run python run_tests.py all

Expected output: ✅ ALL PHASES PASSED