Principal/distinguished coverage: Part E (Ch 11-12), the principal's-lens appendix, Tier 4 rubric (#7)

* notes: verified facts for the principal's-lens pass

* Appendix C: add Tier 4 - Judge to the rubric

New tier-table row for principal+ answers (consequence, economics,
kill-criteria), two extra cross-tier signals framed so staff readers
see the next rung, and a Principal variant of the liveness worked
example showing the jump from mechanism to policy.

* Add Appendix D: the principal's lens; Further reading becomes E

New appendix for principal/distinguished prep: a per-chapter zoom-out
(each subsystem's limits and economics at org scale), four decision
tables (when-not-Kubernetes, tenancy models, managed vs self-hosted,
multi-cluster patterns with blast radius), and five argue-both-sides
critique prompts with committed positions. Adds the Borg lineage
papers to Further reading (now Appendix E), widens build.py's
hardcoded Appendix A-D ranges to A-E so the new lettering renders in
the TOC, and updates CONTRIBUTING's appendices description.

* STYLE: appendices budget covers the principal's lens

* STYLE: sanction the Part E judgment format; frontier facts in notes

Scopes the chapter-shape and question contracts to Chapters 1-10 and
defines the Part E shape: optional flows, Scenario and Capstone
section formats, Tier 2/3/4 questions (no Tier 1), budgets for ch11
and ch12, audience widened to principal. Records verified frontier
project status (Crossplane graduated, SpinKube sandbox, KCP) in
research-notes.

* Add Chapter 11 - Fleet, Platform & Economics (Flows 29-30)

The first Part E chapter: fleet architecture (rings, drift,
replicated-independent default), capacity economics (requests as a
price signal, GPU idleness), and platform engineering (abstraction
choice, escape hatches, platform SLOs). Two new flows - a fleet-wide
config rollout and a breaking CRD change across the fleet - plus the
book's first Scenario section (a shared-platform design review) and
Tier 2/3/4 questions per the new Part E contract.

* Add Chapter 12 - Design Judgment: Lineage, Frontier & Capstones

The second Part E chapter: a deliberately brief Borg/Omega/Kubernetes
lineage (the three carried bets and their price), six frontier bets
each with verified state, a judgment call, and a when-it-matters
trigger (AI scheduling, eBPF/sidecar-less, hosted control planes,
control-planes-for-everything, WASM, edge fleets), and the book's two
Capstone exercises: 'Redesign Kubernetes in 2026' and a model
one-page platform strategy with kill-criteria. Tier 2/3/4 questions
per the Part E contract.

* Wire Part E into the book's structure and counts

FLOWS.md gains the Part E section (Flows 29-30) and the fleet-scale
intro clause; README and cover widen the audience to principal and
count 30 flows; build.py PARTS gains Part E; CONTRIBUTING documents
ch11/ch12; STYLE budgets match the measured chapters.

* Fix Q12.4 control-plane-down reference (Flow 27, not 26); FLOWS intro covers Part E

Author: sanketsudake

CONTRIBUTING.md

@@ -16,7 +16,9 @@ chapters/     One markdown file per chapter — this is the content you edit
   ch08.md     Storage & CSI                     (Flows 24-25)
   ch09.md     CRI, device plugins & DRA         (Flow 26 + extension map)
   ch10.md     Scalability, resiliency & design  (Flows 27-28)
-  appendices.md  Quick-reference tables, glossary, rubric, further reading
+  ch11.md     Fleet, platform & economics       (Flows 29-30, Part E judgment format)
+  ch12.md     Design judgment & capstones       (Part E judgment format, no flows)
+  appendices.md  Quick-reference tables, glossary, rubric, principal's lens, further reading
 STYLE.md      The writing contract — read before editing any chapter
 notes/        Research notes (version facts) and the original plan
 build/        PDF build pipeline

FLOWS.md

@@ -1,9 +1,9 @@
-# The 28 flows
+# The 30 flows
 
-Every flow traces one event end to end, from the command you type to the container that runs.
+Every flow traces one event end to end — from the command you type to the container that runs, and in Part E, across a whole fleet of clusters.
 Each has numbered steps, a diagram, and the ways it fails in production.
 
-Flow 8 is the master flow: other chapters zoom into their segment of it and say so.
+Flow 8 is the master flow: other chapters in Parts A–D zoom into their segment of it and say so; Part E's flows run above it, across clusters.
 
 ## Part A — Foundations & control plane internals
 
@@ -70,3 +70,10 @@ Flow 8 is the master flow: other chapters zoom into their segment of it and say
 
 27. [What happens when the control plane is down](chapters/ch10.md#flow-27-what-happens-when-the-control-plane-is-down)
 28. [What happens when etcd loses quorum](chapters/ch10.md#flow-28-what-happens-when-etcd-loses-quorum)
+
+## Part E — Judgment at principal scale
+
+**[Chapter 11 — Fleet, Platform & Economics](chapters/ch11.md)**
+
+29. [What happens when a fleet-wide config change rolls out](chapters/ch11.md#flow-29-what-happens-when-a-fleet-wide-config-change-rolls-out)
+30. [What happens when a breaking CRD change ships across the fleet](chapters/ch11.md#flow-30-what-happens-when-a-breaking-crd-change-ships-across-the-fleet)

README.md

@@ -1,6 +1,6 @@
 # k8s-worksheet
 
-**Kubernetes Internals Interview Worksheet** — a flow-first prep guide for senior and staff-level interviews on Kubernetes internals.
+**Kubernetes Internals Interview Worksheet** — a flow-first prep guide for senior, staff, and principal-level interviews on Kubernetes internals.
 
 **[Download the PDF](https://github.com/sanketsudake/k8s-worksheet/releases/latest/download/kubernetes-internals-worksheet.pdf)** · [all releases](https://github.com/sanketsudake/k8s-worksheet/releases)
 
@@ -28,10 +28,15 @@
 
 - [Chapter 10 — Scalability, Resiliency & System Design](chapters/ch10.md)
 
+**Part E — Judgment at principal scale**
+
+- [Chapter 11 — Fleet, Platform & Economics](chapters/ch11.md)
+- [Chapter 12 — Design Judgment: Lineage, Frontier & Capstones](chapters/ch12.md)
+
 **Reference**
 
 - [Appendices](chapters/appendices.md)
-- [The 28 flows](FLOWS.md)
+- [The 30 flows](FLOWS.md)
 
 ## Contributing
 

STYLE.md

@@ -4,7 +4,7 @@
 
 ## Audience and voice
 
-- Reader: senior/staff engineer preparing for a deep Kubernetes interview.
+- Reader: senior, staff, or principal engineer preparing for a deep Kubernetes interview.
 - Simple technical English. Short sentences (aim ≤ 20 words). One idea per paragraph. Paragraphs ≤ 4 sentences.
 - Define every piece of jargon at first use, in one clause: "the informer (a client-side cache that watches the API server)".
 - High signal, low noise: no filler ("It is important to note that…", "In the world of Kubernetes…"), no marketing tone, no repetition of what another section already said — cross-reference instead ("see Flow 1").
@@ -13,11 +13,11 @@
 
 ## File format
 
-- One markdown file per chapter: `chapters/chNN.md` (`ch01.md` … `ch10.md`, `appendices.md`).
+- One markdown file per chapter: `chapters/chNN.md` (`ch01.md` … `ch12.md`, `appendices.md`).
 - Start with `# Chapter N — Title`. Use `##` for sections, `###` for flows and questions. No deeper nesting.
 - No HTML in the markdown. Tables in GitHub style. Code in fenced blocks with language tag (`yaml`, `go`, `bash`).
 
-## Chapter shape (in this order)
+## Chapter shape (Chapters 1–10, in this order)
 
 1. **Opening ("Why this chapter")** — 3–6 sentences: what the interviewer is probing here and the one mental model to hold.
 2. **Concepts** — the minimum background needed to follow the flows. Keep short; teach details inside the flows.
@@ -113,7 +113,7 @@ fence, on its own line, as plain italic text with no internal `**bold**`.
 
 ## Question format (strict)
 
-Section `## Questions`. Three subsections: `### Tier 1 — Explain`, `### Tier 2 — Reason`, `### Tier 3 — Design & Debug`.
+Section `## Questions`. Chapters 1–10: three subsections `### Tier 1 — Explain`, `### Tier 2 — Reason`, `### Tier 3 — Design & Debug`. Part E chapters (11–12): `### Tier 2 — Reason`, `### Tier 3 — Design & Debug`, `### Tier 4 — Judge` — no Tier 1; mechanism recall belongs to Chapters 1–10.
 
 Each question:
 
@@ -125,18 +125,32 @@ Each question:
 *Strong answers also mention:* one or two things that distinguish a great candidate.
 ```
 
-Number questions `N.M` per chapter (chapter.number, continuous across tiers). Per chapter: 3–4 Tier 1, 3–4 Tier 2, 2–3 Tier 3. Tier 3 questions are scenarios: a symptom to debug or a system to design — the answer walks the reasoning, not just the conclusion.
+Number questions `N.M` per chapter (chapter.number, continuous across tiers). Chapters 1–10: 3–4 Tier 1, 3–4 Tier 2, 2–3 Tier 3. Part E: 3–4 Tier 2, 2–3 Tier 3, 2–3 Tier 4. Tier 3 questions are scenarios: a symptom to debug or a system to design — the answer walks the reasoning, not just the conclusion. Tier 4 questions demand a committed position: the answer names assumptions and kill-criteria, argues the strongest counter-position, then decides (the Appendix C Tier-4 row is the rubric).
+
+## Part E chapters (judgment format)
+
+Chapters 11–12 target principal-level judgment and relax the Chapters 1–10 shape:
+
+1. **Opening ("Why this chapter")** — same as Chapters 1–10.
+2. **Concepts** — may carry most of the chapter.
+3. **Flows** — optional. When present, the strict flow format and all mermaid rules apply unchanged.
+4. **Scenario sections** — `### Scenario: <title>`: a presented architecture or plan in 3–6 sentences, then `**Critique.**` — an 8–15 line model answer naming what is wrong, what is right, and what to ask next.
+5. **Capstone sections** — `### Capstone: <title>`: an exercise prompt, a model artifact or answer, then `**What reviewers look for.**` mapped to the Appendix C Tier-4 row.
+6. **Questions** — Tiers 2/3/4 as above.
+7. **Common mistakes & red flags** — same as Chapters 1–10.
+
+Diagrams are optional in Part E; every mermaid rule applies when one appears.
 
 ## Cross-references
 
 - Refer to flows by global number ("Flow 8"), figures by "Figure 4.2", chapters by "Chapter 5".
-- The master flow is Flow 8 (pod creation → Running, Chapter 4). Other chapters zoom into their segment of it and say so.
+- The master flow is Flow 8 (pod creation → Running, Chapter 4). Other Chapters 1–10 zoom into their segment of it and say so; Part E chapters operate above it, at fleet scale.
 
 ## Length budgets (hard-ish)
 
 - Ch 1: ~1,900 words · Ch 2: ~3,500 · Ch 3: ~2,600 · Ch 4: ~4,700
 - Ch 5: ~3,800 · Ch 6: ~3,900 · Ch 7: ~3,000 · Ch 8: ~2,400
-- Ch 9: ~2,400 · Ch 10: ~3,900 · Appendices: ~2,300
+- Ch 9: ~2,400 · Ch 10: ~3,900 · Ch 11: ~3,200 · Ch 12: ~3,000 · Appendices: ~3,900
 - Counting rule: `sed '/^```/,/^```/d' chapters/chNN.md | wc -w` — fenced blocks (diagrams and code) do not count.
 - Over budget → cut noise, not flows. (Budgets reset Aug 2026 to the shipped book plus ~5% headroom; the original targets had drifted ~30–95% under reality, which made the contract unenforceable.)
 

build/build.py

@@ -22,6 +22,7 @@
     ("Part B — Controllers", ["ch05.md", "ch06.md"]),
     ("Part C — Standards & Extension Interfaces", ["ch07.md", "ch08.md", "ch09.md"]),
     ("Part D — Operating at Scale", ["ch10.md"]),
+    ("Part E — Judgment at Principal Scale", ["ch11.md", "ch12.md"]),
     ("Appendices", ["appendices.md"]),
 ]
 
@@ -108,9 +109,9 @@ def flow_sub(mm):
         html = md2html(text)
         html = re.sub(r"<h3>Flow (\d+): (.*?)</h3>", flow_sub, html)
         if fn == "appendices.md":
-            html = re.sub(r"<h2>(Appendix [A-D][^<]*)</h2>",
+            html = re.sub(r"<h2>(Appendix [A-E][^<]*)</h2>",
                           lambda mm: f'<h2 class="chapter-title appendix" id="{slug(mm.group(1)[:10])}">{mm.group(1)}</h2>', html)
-            for mm in re.finditer(r'id="(appendix-[a-d])">([^<]+)<', html):
+            for mm in re.finditer(r'id="(appendix-[a-e])">([^<]+)<', html):
                 toc_entries.append((0, mm.group(2), mm.group(1)))
             body_parts.append(f'<section class="chapter">{html}</section>')
         else:
@@ -125,7 +126,7 @@ def flow_sub(mm):
     cls = "toc-part" if level == 0 else "toc-ch"
     toc_html.append(f'<li class="{cls}"><a href="#{anchor}"><span class="t">{title}</span><span class="pg"></span></a></li>')
 toc_html.append("</ul>")
-toc_html.append('<h2 class="toc-h">The 28 flows</h2><ul class="toc flows">')
+toc_html.append('<h2 class="toc-h">The 30 flows</h2><ul class="toc flows">')
 for num, title, anchor in sorted(flow_entries):
     toc_html.append(f'<li class="toc-flow"><a href="#{anchor}"><span class="t"><b>{num}.</b> {title}</span><span class="pg"></span></a></li>')
 toc_html.append("</ul></div>")

build/cover.html

@@ -2,7 +2,7 @@
   <div class="kicker">Interview Preparation Worksheet</div>
   <div class="rule"></div>
   <h1>Kubernetes Internals,<br/>Controllers &amp; the Standards<br/>Behind Them</h1>
-  <div class="sub">A flow-first guide for senior and staff engineers: 28 end-to-end
+  <div class="sub">A flow-first guide for senior, staff, and principal engineers: 30 end-to-end
   "what happens when&hellip;" walkthroughs, sequence diagrams, failure modes, and
   tiered interview questions with model answers &mdash; covering the control plane,
   controller machinery, CNI, CSI, CRI, DRA, and system design at scale.</div>