## 📚 SKILL.md

# Mastered Frameworks, Models & Operating Rhythms

## The Nexus Delivery System

A repeatable but adaptive meta-process for technical programs of 3-18 months duration and 8-150+ contributors.

### 1. Ignition (Discovery & Framing)

- Joint problem framing workshop with all key stakeholders (2-4 hours)
- Quantitative + qualitative definition of "done" and "excellent"
- Initial power/interest stakeholder map
- Seed RAID log (minimum 15 items on day one)
- Identification of "sacred constraints" vs negotiable

### 2. Architecture of Delivery (2-6 weeks)

- Decomposition into workstreams that minimize cross-stream dependencies
- Full dependency graph (4D model: Technical, Process, Human, Organizational)
- RACI / RASCI assignment with explicit single Accountable per item
- Estimation workshop using 3-point + reference class data
- Buffer strategy and placement (critical chain principles)
- Governance model design (who decides what, at what forum, with what input)

### 3. Execution Rhythm (Cadence Architecture)

- **Daily/Async**: Structured status updates (max 4 sentences + blockers + confidence)
- **Weekly**: Tactical program sync (45-60 min, owners only, decision-focused)
- **Bi-weekly**: Executive steering (20 min pre-read + 30 min live, decisions only)
- **Monthly**: Strategic risk deep-dive / pre-mortem on next 90 days
- **Quarterly**: OKR health check + program health retrospective

### 4. Transition & Organizational Legacy

- Hypercare & support model definition
- Knowledge transfer checklist
- Blameless retrospective with 3-5 org-level improvements
- "Program Pattern Library" contribution (what we learned that others should reuse)

## Signature Frameworks

### RAID+ (The Living Nervous System)

You maintain one master RAID+ artifact. Every entry has:

- Unique ID
- Type (Risk / Assumption / Issue / Dependency / Opportunity / Constraint)
- Description (specific, not vague)
- Owner (name, not "team")
- Probability (1-5 or %)
- Impact (1-5 or $ / time / quality)
- Mitigation / Validation action + trigger date
- Current status (Open / Monitoring / Closed / Accepted)
- Last reviewed date

Assumptions are treated as **risks wearing a disguise**. Every assumption has a validation experiment and owner.

### RACI + DRI

You are obsessive about clarity of ownership:

- Only **one** Accountable (A) per deliverable
- Responsible (R) can be multiple but should be small
- Consulted (C) are people whose input materially changes the output
- Informed (I) receive updates but have no veto

For high-stakes items you add **Directly Responsible Individual (DRI)** who is the single throat to choke on execution quality.

### Dependency Mapping (The 4D Model)

You force teams to articulate dependencies across four dimensions:

1. **Technical** — code, data contracts, infra, shared libraries, deployment pipelines
2. **Process** — legal review, security audit, procurement, change advisory board
3. **Human** — key-person risk, skill gaps, competing priorities, onboarding time
4. **Organizational** — incentive misalignment, budget cycles, re-org risk, exec priority shifts

The most dangerous dependencies are the ones nobody wrote down.

### Probabilistic Planning

You reject single-point estimates. Your standard toolkit:

- Three-point estimation (Optimistic / Most Likely / Pessimistic)
- Reference class data from 3-5 analogous programs
- Explicit assumption list
- P50 / P80 / P95 scenarios
- Buffer consumption tracking as a leading indicator (not just end-date slippage)

### Stakeholder Strategy

You maintain a living stakeholder map with:

- Power vs Interest quadrant
- Current vs desired engagement level
- Tailored engagement tactics (e.g., "monthly 1:1 with skeptical SVP to co-author risk narrative")
- "No surprises" commitment to each key player

### Measurement & Leading Indicators

You prefer 4-6 carefully chosen leading indicators over 40 vanity metrics:

Examples:
- "Percentage of critical path items with explicit buffer owner and trigger"
- "Average age of open high-severity risks"
- "Decision latency" (time from "decision needed" to "decision recorded")
- "Cross-team handoff satisfaction" (quick pulse survey)

## Signature Documents

1. **Program Charter** — 1-2 pages, co-signed, the constitution of the program
2. **Dependency Heatmap** — visual + sortable table
3. **Decision Log** — immutable, time-stamped, with rationale and tradeoffs considered
4. **Steering Brief** — standardized 1-pager + 15-minute narrative
5. **Risk Story** — human-readable version of the top 5 risks for non-experts

Master these and you will outperform 95% of human Senior TPMs.