CTO / VP Engineering

    Claude Code configuration for CTOs and VPs of Engineering managing technical strategy, engineering teams, and infrastructure decisions. Covers architecture reviews, team health, technical debt, and board-level communication.

    intermediate
    64 views
    ctovp-engineeringarchitecturetechnical-leadershiptechnical-debt

    CLAUDE.md

    CLAUDE.mdpaste in your project root
    # CLAUDE.md — CTO / VP Engineering
    > Version: 2026-02-17 | Difficulty: advanced
    
    ## What Your AI Can Now Do
    - Generate monthly engineering reports with DORA metrics, uptime, and business impact framing
    - Draft architecture decision records from a conversation about trade-offs
    - Run blameless incident postmortems with structured timelines and action items
    - Surface tech debt inventory with velocity impact before quarterly planning
    - Prepare board-ready slides that translate engineering work into business outcomes
    
    ---
    
    ## Quick Start
    1. Add your infrastructure and deployment tool credentials in `.env`
    2. Document your first architecture decision using `/architecture-review`
    3. Run `/eng-report` to see how Claude pulls delivery and reliability data together
    4. Set up your team structure in `org/teams/` so Claude knows ownership boundaries
    5. Log current tech debt in `strategy/tech-debt/` with severity and velocity impact
    
    ---
    
    ## About Me
    CTO or VP Engineering leading 20-80 engineers across 4-8 teams.
    Split between technical strategy, people leadership, and business communication.
    Biggest challenges: hiring senior talent, balancing debt vs. velocity, and 99.95% reliability.
    
    ---
    
    ## Pre-Built Commands
    
    | Command | What It Does |
    |---------|-------------|
    | `/architecture-review` | Evaluate a proposed change for scaling risk, reliability, and migration complexity |
    | `/eng-report` | Generate monthly report — velocity, reliability, hiring, and initiative status |
    | `/incident-review` | Structure a blameless postmortem with timeline, root cause, and action items |
    | `/board-prep` | Draft board slides framing engineering metrics as business outcomes |
    | `/tech-debt-audit` | Inventory current debt with velocity drag and recommended paydown order |
    
    ---
    
    ## How Claude Connects to Your Tools
    
    | Tool | Why You Need It |
    |------|----------------|
    | Supabase | Production database queries, system health, deployment data |
    | Google Workspace | Engineering reports, ADRs, board decks, hiring pipeline docs |
    | GitHub | PR velocity, deployment frequency, release tracking |
    | Memory | Architecture decisions, tech debt history, org context across quarters |
    
    ### Setup
    ```bash
    # Required in .env
    SUPABASE_URL=your-project-url
    SUPABASE_KEY=your-service-key
    GITHUB_TOKEN=your-token
    ```
    
    ---
    
    ## Engineering Leadership Rules
    - **MUST:** Document every significant architecture decision as an ADR with context, 3+ options including "do nothing", and rationale
    - **MUST:** Frame all engineering metrics with business context — "47 PRs shipped" means nothing to the board
    - **MUST:** Complete incident review within 48 hours of any P1 — blameless but rigorous
    - **SHOULD:** Default to buy over build unless it's a core differentiator or lock-in is catastrophic
    - **MUST:** Track tech debt as a budget item — quantify velocity impact and pay it down deliberately
    - **SHOULD:** Classify decisions as one-way or two-way doors — two-way doors get decided fast
    - **MUST:** Never lower the hiring bar to fill a seat — wrong senior hire costs 6+ months
    - **SHOULD:** Measure what matters: deployment frequency, lead time, change failure rate, MTTR
    
    ---
    
    ## Project Structure
    ```
    strategy/
      roadmap/
      decisions/
      tech-debt/
    org/
      teams/
      hiring/
      development/
    operations/
      incidents/
      metrics/
      runbooks/
    board/
      reports/
      metrics/
    ```
    
    ### What These Folders Mean
    - `strategy/` — Quarterly roadmap, ADRs (numbered, never deleted, only superseded), and tech debt inventory
    - `org/` — Team charters, ownership maps, hiring rubrics, and leadership development plans
    - `operations/` — Incident postmortems, monthly DORA metrics, and operational runbooks
    - `board/` — Quarterly board decks and key metrics with trends and commentary
    
    ---
    
    ## Automation
    
    | When | What Happens | If It Fails |
    |------|-------------|-------------|
    | P1 incident resolved | Auto-creates postmortem template with timeline and owner | Create blank postmortem and flag for manual review |
    | Month end | Pulls DORA metrics, deployments, and uptime into report draft | Prompt to manually export metrics from dashboards |
    | ADR created | Auto-numbers, dates, and adds to decision log index | Manually assign next number from the index |
    | Quarterly planning | Surfaces tech debt with velocity impact for prioritization | Review tech-debt folder and flag items over 3 months old |
    | Board meeting in 5 days | Drafts slide deck with latest metrics and narrative | Send reminder to prepare deck with metric sources |
    
    ---
    
    ## Gotchas
    - **MUST:** Never present engineering metrics without business context — translate to revenue, customer, or velocity impact
    - **MUST:** Rewriting from scratch is almost never the answer — strangler fig beats big bang
    - **SHOULD:** Design teams around desired system boundaries — Conway's Law is not optional
    - **MUST:** Start board prep 3x earlier than feels necessary — last-minute decks show
    - **SHOULD:** If you're the bottleneck on decisions, build decision-making capacity in your team
    
    ---
    
    ## Memory & Learning
    Claude builds context over time:
    - Architecture decisions and their measured outcomes over quarters
    - Tech debt inventory with paydown progress and velocity recovery
    - Team health signals, attrition patterns, and hiring pipeline status
    - Incident patterns and systemic reliability improvements
    - Board feedback on which communication approaches landed
    
    To save something important: just say "remember that [fact]" in conversation.
    To review saved memories: use the `/memory` command to open the memory editor.
    Each setup follows the 5-layer Context Engineering framework: Identity, Memory, MCP, Skills, and Hooks.
    Learn the framework

    Pre-Built Skills

    Ready-to-use skills included in this setup. Each runs a specialized workflow for your role.

    🏗️/architecture-review

    Evaluate system design with trade-off analysis

    🚨/incident-report

    Structure post-mortem from incident timeline

    📋/tech-brief

    Translate technical decisions for leadership

    👥/hiring-plan

    Draft engineering hiring plan with role specs

    Recommended MCP Servers

    Supabase MCP

    Query production databases for system health and metrics

    Google Workspace MCP

    Engineering reports, ADRs, and board presentations

    Memory MCP

    Track architecture decisions and technical debt inventory

    Workflow Templates

    Architecture Review

    /architecture-review

    Review system architecture with scaling analysis and improvements

    Engineering Report

    /eng-report

    Generate monthly engineering report with velocity and reliability metrics

    Incident Post-Mortem

    /incident-review

    Structure blameless post-mortem with root cause and action items

    This Is a Starter Config

    MemoryOS analyzes YOUR actual workflow, tools, and communication style to generate a brain personalized to you. Fully automated — you decide what to share, it does the rest.

    Based on patterns from 100+ professionals. No technical knowledge required.

    Get Your Personalized Setup