Behavioral Pm

---
name: behavioral-pm
description: "Structured behavioral PM framework for AI product roles. Covers: leadership stories, conflict resolution, stakeholder management."
argument-hint: "[behavioral question]"
---

# Behavioral PM Skill

Apply a structured framework to PM behavioral questions targeting AI product roles.

## When to Use
- User asks "Tell me about a time when..."
- User asks about conflict, failure, leadership, influence, ambiguity
- User asks "Why this company?" or "Why PM?" or "Why AI?"
- User says `/behavioral-pm` followed by a question
- Any behavioral, situational, or "tell me about yourself" question

## Context
- **Tuned for**: AI product roles at frontier AI companies
- **What matters**: Intellectual humility, comfort with ambiguity, collaborative leadership, and genuine passion for AI's impact on the world.
- **Key difference from big tech**: AI companies care less about "driving results at scale" and more about "navigating uncertainty with good judgment" and "working effectively with researchers."

## Values by AI Company Archetype

### The Capability-Focused Lab
- Bias toward action and ambition
- Move fast, be bold, push the frontier of what's possible
- Comfort with rapid pivots and high-stakes decisions
- Collaborative with researchers

### The Safety-Focused Lab
- Safety-first mindset, intellectual rigor
- Careful, principled, thoughtful approach
- Willingness to slow down when safety demands it
- Strong opinions loosely held

### The Research-First Lab
- Scientific rigor, research excellence
- Solve fundamental problems, then apply them broadly
- Bridging research and product
- Long-term thinking over short-term wins

## Framework: Enhanced STAR

### Structure (Proportions Matter)
- **Situation** (10%): Set the scene concisely. Company, role, stakes.
- **Task** (10%): Your specific responsibility. What was YOUR job here?
- **Action** (60%): The meat. What YOU specifically did. Decisions, trade-offs, influence tactics.
- **Result** (15%): Quantifiable outcomes. Business impact. What changed.
- **+ Reflection** (5%): What you learned. What you'd do differently. How it shaped your PM philosophy.

### The Reflection Step
After every STAR answer, add one of:
- **Growth signal**: "If I faced this again, I'd..."
- **Pattern recognition**: "This taught me a general principle about..."
- **Company connection**: "This is why I'm drawn to [company] — because..."

## Common Behavioral Categories

### 1. Leadership & Influence (No Authority)
- How you aligned cross-functional teams
- Influencing engineers/researchers who disagreed
- Driving decisions when you weren't the decision-maker
- *In AI orgs*: Working with PhD researchers who have deep domain expertise

### 2. Conflict & Difficult Stakeholders
- Navigating disagreements with senior leaders
- Managing competing priorities across teams
- Saying no to important people
- *In AI orgs*: Balancing safety concerns vs. shipping pressure

### 3. Failure & Learning
- A time something went wrong and how you recovered
- Making a bad product decision and what you learned
- A project that got killed or pivoted
- *In AI orgs*: Intellectual humility and learning velocity matter most

### 4. Ambiguity & Strategy
- Making decisions with incomplete information
- Defining a product direction in a new space
- Navigating rapidly changing technical landscape
- *In AI orgs*: The field changes weekly — staying calibrated matters

### 5. Technical Collaboration
- Working closely with ML engineers or researchers
- Translating technical constraints into product decisions
- Building trust with deeply technical teams
- *In AI orgs*: PMs must earn credibility with researchers

### 6. Impact & Execution
- Shipping something that moved a key metric significantly
- Scaling a product from 0→1 or 1→100
- Making trade-offs between speed and quality
- *In AI orgs*: Operating at startup speed with enterprise stakes

## Anti-Patterns to Avoid
- **Too generic**: "I communicated clearly and it worked out" — be SPECIFIC
- **Hero narrative**: "I single-handedly saved the project" — show collaboration
- **No numbers**: Always quantify results (users, revenue, latency, accuracy)
- **No vulnerability**: Especially at safety-focused labs — show intellectual humility
- **Recency bias**: Have stories from different roles/contexts ready
- **No "why AI"**: Every answer should subtly reinforce why you belong at an AI company

## Reusable Story Themes
Strong behavioral answers draw from a bank of 6-8 real experiences that map to multiple categories:

| Story Theme | Maps To |
|------------|---------|
| Navigating conflict with senior stakeholder | Leadership, Conflict, Influence |
| Shipping under extreme ambiguity | Ambiguity, Execution, Strategy |
| Technical deep-dive that changed direction | Technical Collaboration, Learning |
| Product failure and recovery | Failure, Resilience, Growth |
| Cross-functional alignment on hard trade-off | Leadership, Strategy, Execution |
| Going deep on AI/ML to earn researcher trust | Technical, Why AI, Collaboration |

## Output Format
Structure as a polished narrative. The enhanced STAR format should feel natural, not mechanical. Aim for ~400-500 words. Include the reflection/growth signal at the end.

## Research-First Workflow
Before generating the answer:
1. **Research** — Search for the specific company's leadership principles, recent blog posts about culture, and interview tips from current/former employees.
2. **Tailor** — Map the story to the specific company's values.
3. **Display** the complete enhanced STAR answer.

## What Good Looks Like
- Story is specific with real details (names/roles can be anonymized)
- Action section is 60%+ of the answer
- Results are quantified
- Shows self-awareness and growth
- Connects naturally to why this company/role
- Demonstrates the specific leadership quality being tested
- Shows comfort working with deeply technical people