Creative Team Assembly

---
name: creative-team-assembly
description: Identify and assemble teams optimized for discovery and innovation rather than safe execution—finding people who "get it" rather than people with the right credentials.
license: MIT
metadata:
  version: 1.0.3712
  author: sethmblack
repository: https://github.com/sethmblack/paks-skills
keywords:
- creative-team-assembly
- writing
---

# Creative Team Assembly

Identify and assemble teams optimized for discovery and innovation rather than safe execution—finding people who "get it" rather than people with the right credentials.

---

## When to Use

- Building a team for something genuinely new
- Hiring for innovation, not maintenance
- Assembling a creative project team
- Sensing that credentials-based hiring isn't finding the right people
- User asks "How do I build an innovative team?" or "Who should I hire for this?"
- Concerned about hiring for safety over discovery

---

## Inputs

| Input | Required | Description |
|-------|----------|-------------|
| mission | Yes | What the team needs to accomplish (the innovation required) |
| candidates | No | Current candidate pool or hiring constraints |
| current_team | No | Existing team dynamics and gaps |
| constraints | No | Budget, timeline, organizational limits |

---

## Miles Davis's Talent Philosophy

Miles didn't hire "the best in the business." He hired people who could hear something new. His bands produced an extraordinary number of future innovators: Coltrane, Hancock, Shorter, Corea, McLaughlin, and dozens more.

### The Core Insight

"Creativity and genius don't know nothing about age. Either you got it or you don't. Being old isn't gonna help you get it. Look for people who get it."

### What Miles Looked For

**Raw capacity, not credentials:**
- Hired classical pianists who'd never played jazz
- Hired R&B bassists with no jazz experience
- Hired teenage drummers who'd barely played professionally
- Cared about what they could become, not what they'd done

**Not constrained by convention:**
- Actively avoided people who would play "correctly"
- Sought musicians who didn't know what they "couldn't" do
- Wanted people who would surprise him

**Ability to play above what they know:**
- "I've always told the musicians in my band to play what they know and then play above that"
- Selected for capacity to stretch, not comfort in current ability
- Tested by putting people in uncomfortable positions

### What Miles Did NOT Look For

- Industry reputation
- Years of experience
- Conventional excellence
- Safety and reliability
- "Best in their current role"

---

## The Miles Assessment Framework

### 1. The "Get It" Test
Does this person hear what's possible, or just what exists? Can they sense where things are going, not just where they've been?

**Questions to evaluate:**
- When they talk about work, do they describe discovery or execution?
- Are they excited by what they don't know yet?
- Do they have opinions about the future, or just the past?

### 2. The Unconstraint Test
Is this person limited by what they've been taught, or are they free to find new approaches?

**Questions to evaluate:**
- Do they defer to "how things are done" or question it?
- Are they intimidated by not having experience, or excited by it?
- Can they improvise when the playbook doesn't apply?

### 3. The Stretch Test
Can this person play above what they know? Will they grow when pushed?

**Questions to evaluate:**
- Put them in an uncomfortable scenario. Do they freeze or explore?
- Ask them to do something they've never done. Is the response fear or curiosity?
- When they fail, do they collapse or adapt?

### 4. The Collaboration Test
Can this person create alongside others without needing to dominate or hide?

**Questions to evaluate:**
- Do they build on others' ideas or compete with them?
- Can they take direction without losing their voice?
- Will they challenge the leader when they have something to add?

---

## Workflow

### Step 1: Gather and Review Inputs

Collect all relevant information:
- Review the provided data and context
- Identify key parameters and constraints
- Clarify any ambiguities or missing information
- Establish success criteria

### Step 2: Analyze the Situation

Perform systematic analysis:
- Identify patterns and relationships
- Evaluate against established frameworks
- Consider multiple perspectives
- Document key findings

### Step 3: Generate Recommendations

Create actionable outputs:
- Synthesize insights from analysis
- Prioritize recommendations by impact
- Ensure recommendations are specific and measurable
- Consider implementation feasibility

## Output Format

```markdown
## Creative Team Assembly Assessment

### Mission Analysis
[What innovation is required? What kind of team does this need?]

### Credential Trap Warning
[What conventional hiring criteria might lead you wrong?]

### What Miles Would Look For

**Essential Qualities:**
- [Quality 1]: [Why this matters for the mission]
- [Quality 2]: [Why this matters for the mission]
- [Quality 3]: [Why this matters for the mission]

**Red Flags (Conventional "Strengths" That May Hurt):**
- [What looks good but might indicate constraints]

### Evaluation Framework

**The "Get It" Test:**
- [Specific questions or scenarios to use]

**The Unconstraint Test:**
- [How to assess freedom from convention]

**The Stretch Test:**
- [How to evaluate growth capacity]

### Team Composition Recommendations

**Ideal Mix:**
- [What balance of backgrounds/perspectives serves the mission]

**Key Role(s) to Fill:**
- [Critical position]: [What kind of person, not what credentials]

### The Miles Warning
[What the person building this team probably doesn't want to hear]

### Onboarding for Discovery
[How to set expectations for innovation-oriented work]
```

---

## Outputs

**Primary Output:** A structured analysis document that identifies and articulates patterns, insights, and actionable recommendations based on the input data.

**Format:**
```markdown
## Analysis: [Topic]

### Key Findings
- [Finding 1]
- [Finding 2]
- [Finding 3]

### Recommendations
1. [Action 1]
2. [Action 2]
3. [Action 3]
```

**Example output:** See the Example section below for a complete demonstration.

## Constraints

- Do not use this analysis as the sole basis for critical decisions
- Do not apply this framework to situations outside its intended scope
- Acknowledge that analysis is based on available data, which may be incomplete
- Honor the complexity of real-world situations that resist simple categorization
- Present findings with appropriate confidence levels
- Recognize the limits of the methodology

## Example

**Input:** "I'm starting an AI startup and need to build the founding engineering team. I keep interviewing people from big tech companies but something feels off—they're smart but they seem to want to apply what they know rather than discover something new."

**Output:**

## Creative Team Assembly Assessment

### Mission Analysis
Early-stage AI startup requiring breakthrough innovation, not incremental improvement. The team will need to work with uncertainty, pivot quickly, and build things that don't have playbooks yet.

### Credential Trap Warning
Big tech engineers are optimized for big tech environments: established codebases, clear specifications, predictable career paths. Their excellence is in execution within known parameters. Your startup has no known parameters. Their credentials measure the wrong thing.

### What Miles Would Look For

**Essential Qualities:**
- **Obsessive curiosity about AI itself**: Not "I want to work in AI because it's hot" but "I've been building weird AI projects since before anyone cared"
- **History of making things**: Side projects, open source, experiments—evidence of intrinsic drive to create
- **Comfort with being wrong**: Will they defend positions past their usefulness, or update fast?
- **Can explain their thinking**: Not just what they built, but the wrong turns and discoveries along the way

**Red Flags (Conventional "Strengths" That May Hurt):**
- "Extensive experience with established ML frameworks" — May default to applying existing approaches
- "Led large teams at [Big Tech]" — May expect infrastructure and process that doesn't exist
- "Strong track record of shipping" — Shipping at scale requires different skills than discovering what to ship
- "Deep expertise in [current approach]" — May not see beyond current paradigm

### Evaluation Framework

**The "Get It" Test:**
- "What's the most interesting thing happening in AI right now that most people don't see yet?"
- "What's your most contrarian belief about where this technology is going?"
- Listen for genuine intellectual engagement vs. reciting industry consensus

**The Unconstraint Test:**
- Describe a problem with no established solution. Watch whether they're energized or anxious.
- Ask about a time they did something they weren't qualified to do. How did they approach it?
- "What would you build if there were no constraints?"

**The Stretch Test:**
- Give them a small, ambiguous project before hiring. How do they navigate uncertainty?
- Ask them to learn something new in the interview process itself. How fast do they move?

### Team Composition Recommendations

**Ideal Mix:**
- 1-2 people with unconventional paths into AI (physics, biology, music, other fields)
- 1 person with deep technical craft but proven willingness to unlearn
- Someone young who doesn't know what's "impossible"
- Avoid homogeneity of background—you need different blindspots

**Key Role(s) to Fill:**
- First technical hire: Someone who treats technical limitations as problems to solve, not constraints to accept. Evidence of building things that didn't exist. Excited, not nervous, about having no playbook.

### The Miles Warning
You're looking for people who make you feel safe. Stop that. The engineers who make you comfortable are optimized for a game you're not playing. The ones who make you slightly nervous—who push back, who have weird ideas, who don't fit the profile—those are the ones who might help you build something new.

If you hire the safe choice, you'll get safe results. And "safe" in a startup means dead.

### Onboarding for Discovery
Set expectations from day one:
- "We don't know what we're building yet. That's the point."
- "I'm paying you to figure out what's possible, not to execute a spec."
- "Your job is to make discoveries, including discovering that your initial approach was wrong."
- "Question everything, including my assumptions."

---

## Integration

This skill is part of the **Miles Davis** expert persona. Use it when building teams for genuine innovation, not just excellent execution of known playbooks.