Generate Story Bank

Generate Story Bank

Trigger phrases: "story bank", "interview prep", "behavioral stories", "HPARL", "interview stories"

HPARL vs STAR

The key insight: STAR tells what happened. HPARL shows how you think.

The HPARL Format

1. Hook (10-15 seconds)

Promise the interviewer you'll answer their question AND make them want to listen.

Bad hook: "At Anchorage, I worked on ETH staking infrastructure." Good hook: "I'll tell you about the time we had $2B in staked ETH and I had to ensure we never lost a single dollar to slashing — while shipping 8 new protocols in parallel."

Formula: Stakes + constraint + tease of outcome

2. Principles (30 seconds)

Your philosophy or approach that guided the work. This is where you show judgment.

Example: "My principle with infrastructure products is that reliability is the feature — users don't notice when things work, they only notice when things break. So I always over-invest in redundancy before we need it, not after we've had an incident."

Formula: "I believe..." or "My approach to X is..." + why

3. Action (60-90 seconds)

What YOU specifically did. Use "I", not "we". Be concrete.

Structure:

• First, I...
• Then, I...
• The key decision was...

Include:

• Concrete actions you took
• Decisions you made (and why)
• Trade-offs you navigated
• How you worked with others (but your role)

4. Results (30 seconds)

Quantified outcomes. Multiple metrics if possible.

Structure:

Primary metric: [X outcome]
Secondary metric: [Y outcome]
Business impact: [What it meant for the company]

Example: "Zero slashing events across 18 months. Galaxy, Grayscale, and other institutions trusted us with their staking. We grew from $X to $Y in staked assets."

5. Learnings (15-30 seconds)

What you took away. Shows self-awareness and growth mindset.

Formula: "The counterintuitive insight was..." or "What I learned was..."

Example: "The counterintuitive insight was that slower deployments meant faster growth. Institutions needed to see months of perfect operation before committing significant capital. Velocity isn't always the right metric."

Story Bank Target

You need 10-15 stories that cover different behavioral question types:

Generation Process

Step 1: Select Achievement

First, check competency coverage to understand what stories you need:

npm run check:coverage

This shows which of the 7 PM bundles are covered vs gaps. Target stories that fill gaps.

Search for achievements by theme:

# Find leadership stories
npm run search:evidence -- --terms "led,managed,cross-functional,team"

# Find technical stories
npm run search:evidence -- --terms "architecture,api,system,infrastructure"

# Find impact/growth stories
npm run search:evidence -- --terms "revenue,growth,launched,shipped"

Or list all achievements:

ls content/knowledge/achievements/

Choose achievements that:

• Have strong quantified outcomes
• Demonstrate judgment (decisions, trade-offs)
• Show your personal contribution
• Cover different story categories
• Fill gaps identified by check:coverage

Step 2: Transform to HPARL

For each achievement, generate:

# content/knowledge/stories/{achievement-id}.hparl.yaml
id: "{achievement-id}-hparl"
sourceAchievement: "{achievement-id}"
format: "HPARL"
lastUpdated: "YYYY-MM-DD"

# ───────────────────────────────────────────────────────────────
# THE STORY
# ───────────────────────────────────────────────────────────────
hook: |
  [10-15 second attention grabber with stakes and constraint]

principles: |
  [Your philosophy/approach that guided this work]

action:
  - "[First concrete action you took]"
  - "[Second action - key decision point]"
  - "[Third action - how you worked with others]"
  - "[Fourth action - trade-off you navigated]"

results:
  - metric: "[Primary outcome]"
    context: "[Why this matters]"
  - metric: "[Secondary outcome]"
    context: "[Business impact]"

learnings: |
  [The counterintuitive insight or what you'd do differently]

# ───────────────────────────────────────────────────────────────
# INTERVIEW MAPPING
# ───────────────────────────────────────────────────────────────
goodFor:
  - "[Question type 1 this answers well]"
  - "[Question type 2]"
  - "[Question type 3]"

categories:
  - leadership          # Which category bucket
  - technical-challenge

# ───────────────────────────────────────────────────────────────
# PRACTICE NOTES
# ───────────────────────────────────────────────────────────────
timing:
  hook: "15s"
  principles: "30s"
  action: "90s"
  results: "30s"
  learnings: "15s"
  total: "3 minutes"

practiceNotes: |
  [Personal notes on what to emphasize, what to skip, common follow-ups]

Step 3: Validate Coverage

After generating stories, check coverage:

storyBankCoverage:
  leadership:
    count: 3
    stories: ["eth-staking-hparl", "l2-integrations-hparl", "team-conflict-hparl"]
  conflict:
    count: 2
    stories: ["stakeholder-alignment-hparl", "technical-pushback-hparl"]
  failure:
    count: 2
    stories: ["launch-delay-hparl", "wrong-architecture-hparl"]
  technical:
    count: 3
    stories: ["eth-staking-hparl", "ankr-api-hparl", "xbox-blockchain-hparl"]
  impact:
    count: 3
    stories: ["ankr-15x-hparl", "eth-staking-hparl", "xbox-blockchain-hparl"]
  ambiguity:
    count: 2
    stories: ["forte-pivot-hparl", "mempools-startup-hparl"]

  totalStories: 12
  gaps: []  # Categories with <2 stories
  recommendation: "Coverage looks good. Consider adding 1 more failure story."

Example Transformation

Source Achievement (STAR format)

id: eth-staking-zero-slashing
headline: "Zero slashing events across $2B+ in staked ETH"
situation: |
  Anchorage needed ETF-grade staking infrastructure...
task: |
  Lead validator architecture and client partnerships...
action: |
  Designed multi-cloud failover, built compliance dashboard...
result: |
  Zero slashing, Galaxy/Grayscale onboarded, $2B staked...

Generated HPARL Story

id: eth-staking-zero-slashing-hparl
sourceAchievement: eth-staking-zero-slashing
format: HPARL

hook: |
  I'll tell you about the time we had $2B in staked ETH across
  institutional clients like Galaxy and Grayscale, and I had to
  ensure we never lost a single dollar to slashing while shipping
  8 new protocol integrations in parallel.

principles: |
  My principle with infrastructure products is that reliability IS
  the feature. Users don't notice when things work perfectly—they
  only notice when things break. So I always over-invest in
  redundancy before we need it, not after an incident forces us to.

action:
  - "I designed a validator orchestration system across 3 cloud providers with automated failover—if AWS went down, we'd be on GCP in under 30 seconds"
  - "I built a compliance dashboard for institutional clients so they could see real-time attestation health, which became a key sales differentiator"
  - "I partnered with our security team to implement multi-sig key management that satisfied SOC2 requirements while maintaining operational speed"
  - "The hardest trade-off was choosing reliability over velocity—I pushed back on rushing new protocol launches until our monitoring was bulletproof"

results:
  - metric: "Zero slashing events"
    context: "across 18 months of operation with $2B+ staked"
  - metric: "Galaxy, Grayscale, and 5 other institutions"
    context: "trusted us with their ETH staking"
  - metric: "8 new protocols integrated"
    context: "while maintaining zero-incident track record"

learnings: |
  The counterintuitive insight was that slower deployments meant
  faster growth. Institutions needed to see months of perfect
  operation before committing significant capital. What looked
  like over-engineering to internal stakeholders was actually
  the fastest path to enterprise trust.

goodFor:
  - "Tell me about a time you built something reliable"
  - "How do you handle pressure to ship faster?"
  - "Describe a technical leadership challenge"
  - "Tell me about working with enterprise clients"

categories:
  - technical-challenge
  - leadership
  - impact

timing:
  total: "2.5-3 minutes"

Quality Checklist

Before marking a story complete:

• Hook creates genuine curiosity (stakes + constraint + tease)
• Principles show YOUR thinking, not generic PM advice
• Actions use "I" not "we" — your specific contribution is clear
• Results have at least 2 quantified metrics
• Learnings show self-awareness (what you'd do differently OR counterintuitive insight)
• Story maps to at least 2 question types
• Total timing is 2-3 minutes when spoken
• You could defend every claim in a follow-up question

File Locations

Commands

# ═══════════════════════════════════════════════════════════════
# STEP 1: Check what competencies need stories
# ═══════════════════════════════════════════════════════════════
npm run check:coverage           # Show gaps in 7 PM bundles
npm run check:coverage -- --json # JSON output for processing

# ═══════════════════════════════════════════════════════════════
# STEP 2: Search for relevant achievements
# ═══════════════════════════════════════════════════════════════
npm run search:evidence -- --terms "leadership,cross-functional"
npm run search:evidence -- --terms "technical,architecture,api"
npm run search:evidence -- --terms "revenue,growth,impact"

# ═══════════════════════════════════════════════════════════════
# STEP 3: List source files
# ═══════════════════════════════════════════════════════════════
ls content/knowledge/achievements/
ls content/knowledge/stories/*.hparl.yaml