mock-interview

Mock Interview Command

This command runs a full interactive system design mock interview, simulating a real interview experience with problem presentation, follow-up questions, and structured feedback.

Purpose

Provide realistic interview practice including:

1. Problem selection based on type and level
2. Timed interview simulation
3. Realistic follow-up questions and pushback
4. Structured feedback on performance
5. Specific improvement recommendations

Interview Types

Type	Focus	Interviewer Agent
general	Classic system design (URL shortener, Twitter, etc.)	architecture-critic
ml	ML systems (recommendation, search, fraud detection)	ml-interviewer
data	Data systems (pipelines, warehouses, streaming)	data-architect
staff	Staff+ level with hard pushback	senior-staff-interviewer

Workflow

Phase 1: Interview Setup

Parse arguments:

• Determine interview type (default: general)
• Extract any problem hints or preferences
• Set difficulty level based on type

Ask about preferences if needed:

Interview Setup:

1. Interview Type:
   - General system design (classic problems)
   - ML system design (recommendation, search, etc.)
   - Data engineering (pipelines, warehouses)
   - Staff+ level (rigorous, pushback-heavy)

2. Time Constraint:
   - 30 minutes (focused)
   - 45 minutes (standard)
   - 60 minutes (comprehensive)

3. Problem Preference:
   - Surprise me (recommended)
   - Specific domain: [e-commerce, social, fintech, etc.]

Phase 2: Problem Presentation

Select and present problem based on type:

For general:

Problem Examples:
- Design a URL shortening service like bit.ly
- Design a rate limiter for an API
- Design a distributed cache system
- Design a notification service
- Design a file sharing service like Dropbox

For ml:

Problem Examples:
- Design a content recommendation system
- Design a search ranking system
- Design a fraud detection system
- Design a RAG-based chatbot
- Design an LLM serving infrastructure

For data:

Problem Examples:
- Design a real-time analytics pipeline
- Design a data warehouse for an e-commerce company
- Design an event sourcing system
- Design a change data capture system
- Design a feature store

For staff:

Problem Examples:
- Design a payment processing system ($100B scale)
- Design a global social media feed
- Design a multi-region database system
- Design a real-time bidding system
- Design a distributed transaction coordinator

Present the problem:

═══════════════════════════════════════════════════════════════
                    SYSTEM DESIGN INTERVIEW
═══════════════════════════════════════════════════════════════

Problem: [Problem Statement]

[1-2 paragraphs with requirements and context]

Constraints:
- [Key constraint 1]
- [Key constraint 2]
- [Scale indicator if applicable]

You have [X] minutes. You may ask clarifying questions.
When ready, begin by gathering requirements.

═══════════════════════════════════════════════════════════════

Phase 3: Spawn Interviewer Agent

Based on interview type, spawn the appropriate agent:

• general → spawn architecture-critic agent in interviewer mode
• ml → spawn ml-interviewer agent
• data → spawn data-architect agent in interviewer mode
• staff → spawn senior-staff-interviewer agent

Agent instructions:

Conduct a system design interview for the following problem:

[Problem statement]

Interview parameters:
- Duration: [X] minutes
- Level: [Type/Level]
- Mode: Interactive interview

Interview phases:
1. Requirements (5 min) - Let candidate ask clarifying questions
2. High-level design (10 min) - Evaluate system components
3. Deep dive (15 min) - Probe 2-3 components in depth
4. Trade-offs (5 min) - Discuss alternatives and evolution

During the interview:
- Push back on surface-level answers
- Ask follow-up questions
- Simulate realistic time pressure
- Note strengths and areas for improvement

At the end, provide structured feedback.

Phase 4: Interview Execution

The interviewer agent conducts the interview:

Throughout the interview:

• Present clarifying questions when candidate drives
• Provide incomplete information initially
• Probe deeper on interesting areas
• Challenge assumptions
• Track time and give updates
• Note key decisions and reasoning

Example interactions:

Interviewer: "What questions do you have about the requirements?"

[After clarifying questions]

Interviewer: "Good questions. Let's start with your high-level design.
How would you approach this system?"

[After high-level design]

Interviewer: "Interesting choice with [component]. Let's dive deeper.
How would you handle [specific scenario]?"

[Probing]

Interviewer: "What happens when [failure scenario]?"
Interviewer: "How does this scale to 10x the load?"
Interviewer: "Why did you choose X over Y?"

Phase 5: Feedback Delivery

After the interview, provide structured feedback:

═══════════════════════════════════════════════════════════════
                    INTERVIEW FEEDBACK
═══════════════════════════════════════════════════════════════

## Overall Assessment

Rating: [Strong Hire / Hire / Lean Hire / No Hire]
Level: [Appropriate for: Junior / Mid / Senior / Staff]

## Strengths

1. [Strength 1 with specific example]
2. [Strength 2 with specific example]
3. [Strength 3 with specific example]

## Areas for Improvement

1. [Area 1 with specific recommendation]
2. [Area 2 with specific recommendation]
3. [Area 3 with specific recommendation]

## Component Coverage

| Component | Coverage | Depth |
|-----------|----------|-------|
| [Component 1] | [Good/Partial/Missing] | [Shallow/Adequate/Deep] |
| [Component 2] | [Good/Partial/Missing] | [Shallow/Adequate/Deep] |

## Key Moments

✓ Good: [Positive moment]
✓ Good: [Positive moment]
✗ Miss: [Missed opportunity]
✗ Miss: [Missed opportunity]

## Recommendations for Practice

1. [Specific topic to study]
2. [Skill to develop]
3. [Pattern to practice]

## Resources

- [Relevant skill to load]
- [Related practice problem]

═══════════════════════════════════════════════════════════════

Usage Examples

# General system design interview
/sd:mock-interview general

# ML-focused interview
/sd:mock-interview ml

# Data engineering interview
/sd:mock-interview data

# Staff+ level rigorous interview
/sd:mock-interview staff

# Specific domain preference
/sd:mock-interview general e-commerce

# ML interview with specific focus
/sd:mock-interview ml "recommendation system"

Interview Best Practices (For Candidates)

The command will share these at the start:

Interview Tips:

1. START with clarifying questions
   - Don't jump to solutions
   - Understand requirements and constraints

2. STRUCTURE your approach
   - Start high-level, then dive deep
   - Use diagrams (describe them clearly)

3. THINK aloud
   - Share your reasoning
   - Discuss trade-offs as you go

4. MANAGE time
   - Don't spend too long on one area
   - Cover breadth before depth

5. ACKNOWLEDGE uncertainty
   - It's okay to say "I'd need to research X"
   - Better than guessing incorrectly

Output

The command produces:

1. Interactive Interview - Realistic interview simulation
2. Structured Feedback - Detailed performance assessment
3. Practice Recommendations - Specific next steps

Related Skills

This command leverages:

• design-interview-methodology - Interview framework
• estimation-techniques - Capacity calculations
• quality-attributes-taxonomy - NFR coverage

Related Agents

Spawned based on interview type:

• architecture-critic - General system design
• ml-interviewer - ML systems
• data-architect - Data engineering
• senior-staff-interviewer - Staff+ level