Handbook
Appendix: Source Coverage & Gap Analysis
Source Coverage Matrix
Maps every part of the source PDFs to the handbook module(s) that absorb it.
Sources analysed (all pages, via text extraction in _source/):
| ID | File | Pages | Notes |
|---|---|---|---|
| S1 | Expanded_PM_AI_and_Tools_Study_Guide.pdf |
95 | Base source. Lectures 1–55, picture lessons, dictionary, references, 28-chapter tools companion (pp. 73–87), 15 additional PM topics (pp. 88–95). |
| S2 | Complete_Visual_PM_AI_and_Tools_Study_Guide.pdf |
87 | Word-level diff against S1: 0 words appear only in S2. S2 = S1 minus the "Additional PM Topics" section. Treated as fully covered by S1. |
| S3 | PM_Technology_Practical_Companion.pdf |
15 | Found in the same folder (not attached). Its table of contents matches S1 companion chapters 1–28. Not separately mapped. |
| C | Curriculum screenshot (12 Product Core modules) | – | Not received. The 12 module titles from the brief are used and marked provisional until verified against the screenshot. |
Source-quality note. S1 is a set of study notes written from auto-generated lecture transcripts. It repeatedly flags where the instructor's statements were inaccurate (e.g. RAG requires a vector DB; LoRA is a model; Ansoff quadrant mix-ups; RICE "reach − impact = churn"). The handbook keeps those corrections (as "Source-flagged misconceptions") rather than the original claims. Lectures 44 and 54 have no recording, so nothing is claimed about their content.
A. Lecture-to-module map (S1 pp. 1–66)
| Lecture | S1 page | Primary module | Also feeds |
|---|---|---|---|
| L1 Orientation | 1 | M1 | – (course logistics only; no PM content) |
| L2 Intro to PM | 2 | M1 | M6 (career), Interview |
| L3 User understanding & research | 3 | M3 | M1 |
| L4 Favourite product critique | 4 | M3 | Interview (product sense) |
| L5 Journey mapping & user flows | 5 | M3 | M5 |
| L6 Market research with GenAI | 6 | M3 | M2 (TAM/SAM/SOM), M10 |
| L7 Guesstimates | 8 | M3 | Interview (estimation) |
| L8 Product case studies | 9 | M4 | Interview (product design) |
| L9 Competitor analysis | 10 | M2 | M3 |
| L10 Product strategy & BYDP intro | 11 | M2 | M1 |
| L11 Improving YouTube Shorts | 12 | M3 | Cases |
| L12 Development methodologies | 13 | M1 | M9 (Agile, CI/CD) |
| L13 Vision, roadmap, Jira | 15 | M2 | M4, M6 |
| L14 PRDs, Confluence, AI docs | 16 | M6 | M10 |
| L15 Ride-hailing surge pricing | 17 | M8 | M2 (pricing), Cases |
| L16 UI/UX, Figma, AI design | 18 | M5 | M11 |
| L17 Prioritization frameworks | 20 | M4 | – |
| L18 PM career Q&A | 21 | Interview | M1 |
| L19 Product teardown | 22 | M3 | Cases |
| L20 MVP | 23 | M4 | M11 |
| L21 North Star & metrics | 24 | M7 | M2 |
| L22 Interview prep 1 | 26 | Interview | M7 (metric drop RCA) |
| L23 GTM strategy | 27 | M8 | M2 |
| L24 Strategy frameworks (Ansoff, Five Whys, STAR) | 28 | M2 | M4, M6, Interview |
| L25 AARRR | 29 | M7 | M8 |
| L26 Acquisition, activation, monetisation | 30 | M8 | M2 (pricing), M7 |
| L27 Hypothesis-driven dev / A/B | 32 | M7 | M4 |
| L28 Product teardown (Claude) | 33 | M3 | M10 (AI eval) |
| L29 Technical basics | 34 | M9 | – |
| L30 Technical PM mock interview | 35 | Interview | M9 |
| L31 Quick-commerce AOV case | 36 | M7 | Cases |
| L32 Build Your Own Product guidance | 38 | Projects | M4 |
| L33 APIs, integrations, testing | 39 | M9 | – |
| L34 SQL fundamentals | 40 | M7 | M9 |
| L35 Vectors & analytics demo | 41 | M10 | M7 (GA setup) |
| L36 Facebook Groups case | 43 | M8 | Cases |
| L37 What is AI PM | 44 | M10 | M1 |
| L38 Framing AI problems | 45 | M10 | M4 |
| L39 Prompting frameworks | 46 | M10 | M11 |
| L40 AI PRDs & data requirements | 47 | M10 | M6 |
| L41 Dating product case | 49 | M8 | M10, Cases |
| L42 Intro to vibe coding | 50 | M11 | – |
| L43 Vibe-coding an MVP | 51 | M11 | M4 |
| L44 Interview Prep 4 | 52 | Interview | Recording unavailable – not covered |
| L45 RAG and workflow | 52 | M10 | M12 |
| L46 MCP & workflows | 54 | M12 | M9 |
| L47 LLMs & fine-tuning | 55 | M10 | – |
| L48 Agents & prototype | 56 | M12 | M11 |
| L49 Agents part 2 | 58 | M12 | M11 |
| L50 Document-QA prototype | 59 | M10 | M11, M12 |
| L51 AI ethics & orchestration | 60 | M12 | M10 |
| L52 AI PM mock interview 1 | 61 | Interview | M10 |
| L53 AI PM mock interview 2 | 62 | Interview | – |
| L54 Agent competition | 64 | M12 | Recording unavailable – not covered |
| L55 BYDP capstones | 64 | Projects | M12 |
B. Picture lessons, dictionary, references (S1 pp. 66–72)
| Source item | Handbook destination |
|---|---|
| Picture lesson: LLM | M10 §LLM inference + diagram |
| Picture lesson: Vectors & embeddings | M10 §Embeddings |
| Picture lesson: RAG | M10 §RAG pipeline diagram |
| Picture lesson: AI agents & safe actions | M12 §Agent loop |
| Picture lesson: API & MCP | M9 §APIs, M12 §MCP |
| Picture lesson: Product discovery & design | M3, M5 |
| Dictionary (≈20 terms) | Glossary (all terms retained, expanded) |
| Selected references (7) | references/ – links re-verified before publishing |
C. Practical companion (S1 pp. 73–87)
| # | Chapter | Module |
|---|---|---|
| 1 | Think like a PM | M1 |
| 2 | Discovery, interviews, segmentation | M3 |
| 3 | Journeys, flows, UX critique | M3, M5 |
| 4 | Requirements & prioritization | M4, M6 |
| 5 | KPIs & worked funnel | M7 |
| 6 | Experiments | M7 |
| 7 | How websites/apps work | M9 |
| 8 | APIs, JSON, webhooks | M9 |
| 9 | Data, databases, events | M7, M9 |
| 10 | AI / ML / GenAI | M10 |
| 11 | LLMs, tokens, context windows | M10 |
| 12 | Embeddings & similarity | M10 |
| 13 | RAG | M10 |
| 14 | Agentic AI & safe automation | M12 |
| 15 | MCP vs API | M12 |
| 16 | AI evaluation, privacy, security | M10, M12 |
| 17 | Figma for PMs | M5 |
| 18 | Jira | M6, M9 |
| 19 | Confluence | M6 |
| 20 | Claude workflows | M11 |
| 21 | ChatGPT workflows | M11 |
| 22 | Gemini workflows | M11 |
| 23 | Lovable | M11 |
| 24 | GitHub for PMs | M11 |
| 25 | n8n | M12 |
| 26 | Replit | M11 |
| 27 | Automating an existing website safely | M12 |
| 28 | Tool selection & 30-day plan | M11 + Resources |
D. Additional PM topics (S1 pp. 88–95)
| # | Topic | Module |
|---|---|---|
| 1 | Vision, strategy, roadmap | M2 |
| 2 | Goals, OKRs, outcomes | M2 |
| 3 | North Star & metric trees | M7 |
| 4 | Retention, cohorts, activation | M7 |
| 5 | Pricing & value | M2, M8 |
| 6 | Unit economics & contribution margin | M7 |
| 7 | CAC & LTV | M7 |
| 8 | Market sizing & segmentation | M2 |
| 9 | Positioning & competitive analysis | M2 |
| 10 | GTM & launch planning | M8 |
| 11 | Stakeholder alignment & trade-offs | M6 |
| 12 | Delivery, dependencies, tech debt | M4, M9 |
| 13 | Incidents, reliability, monitoring | M9 |
| 14 | Accessibility & inclusive design | M5 |
| 15 | PM templates (problem, experiment, decision, launch) | Templates |
E. Status of the new handbook
All 12 modules, the framework library, case library, projects, interview handbook, templates and glossary are written and built (see PROGRESS.md). Each module file under content/modules/ carries a sources: header naming the lectures and companion chapters it preserves; preserved content appears in green-grey "From your course notes" boxes, so every row above can be audited in the built pages.
| Module | Source lectures / chapters absorbed (from the header of each module file) |
|---|---|
| M1 | L1, L2, L12, L20; Companion 1; Extra 1, 11, 15 |
| M2 | L6, L9, L10, L13, L23, L24; Extra 1, 2, 5, 8, 9, 10 |
| M3 | L3, L4, L5, L6, L7, L8, L11, L19, L28, L36; Companion 2, 3; picture lesson "discovery" |
| M4 | L8, L17, L20, L24, L27, L43; Companion 4; Extra 12 |
| M5 | L5, L16, L27; Companion 3, 17; Extra 14; picture lesson "discovery" |
| M6 | L13, L14, L18, L22, L24, L30, L52, L53; Companion 18, 19; Extra 11, 15 |
| M7 | L21, L25, L26, L27, L31, L34, L35; Companion 5, 6, 9; Extra 3, 4, 6, 7 |
| M8 | L15, L23, L25, L26, L36, L41; Extra 5, 10, 13 |
| M9 | L12, L29, L30, L33, L34; Companion 7, 8, 9, 18, 24; Extra 12, 13 |
| M10 | L35, L37, L38, L39, L40, L45, L47, L50, L51; picture lessons; Companion 10–13, 16 |
| M11 | L42, L43, L48, L49, L50; Companion 20–24, 26, 28 |
| M12 | L46, L48, L49, L50, L51, L54 (unavailable), L55; picture lessons; Companion 14, 15, 25, 27 |
| Interview | L18, L22, L30, L44 (unavailable), L52, L53 |
| Projects | L32, L55 |
Curriculum Gap Analysis
Comparison of the existing study guide (S1) against the 12-module Product Core curriculum in the brief. Module titles are provisional because the curriculum screenshot did not arrive.
Legend — Strong: taught with examples in S1 · Partial: mentioned or taught shallowly · Missing: not in S1 (added as independent material in the handbook, labelled "Independent / verified reference", never as course content).
Module by module
| Module | Strong in S1 | Partial | Missing from S1 (handbook adds) |
|---|---|---|---|
| M1 Fundamentals | PM role & specialisations, MVP, Agile/Scrum/Kanban, career paths | PM vs PO vs BA, PLG vs SLG | PM vs Project Mgr / Designer / PMM comparison tables, product lifecycle (PLC), operating models, outcome vs output, discovery vs delivery, platform vs feature, B2B vs B2C, common mistakes |
| M2 Strategy | Vision/strategy/roadmap, Ansoff, competitor analysis, TAM/SAM/SOM, positioning, pricing basics, OKRs | SWOT (named), North Star | Porter's Five Forces (announced but untaught in L24), PESTLE, STP, value propositions, BMC, Lean Canvas, Blue Ocean, moats / network effects, build-buy-partner, portfolio strategy |
| M3 Discovery | Personas, interviews, journey maps, market research, guesstimates, teardown | JTBD (referenced only), segmentation | Interview scripts, survey bias, Opportunity Solution Trees, affinity mapping, assumption mapping, triangulation, review mining |
| M4 Prioritization | RICE, impact/effort, MoSCoW, Five Whys, MVP, hypothesis dev | Dependencies | Kano, ICE, weighted scoring, Cost of Delay, WSJF, Pareto, fishbone, first principles, systems thinking, risk-adjusted prioritisation, experiment prioritisation |
| M5 UX & Design | UI vs UX, Figma basics, wireframe vs prototype, accessibility, usability testing | Journey vs flow | Nielsen heuristics, Fitts/Hick/Jakob, Gestalt, cognitive load, IA, design systems, responsive/mobile, error & empty states, UX metrics (HEART), handoff |
| M6 Communication | PRDs, decision notes, stakeholder trade-offs, STAR, estimates without blame | Jira/Confluence | Pyramid Principle, PREP, storytelling with data, negotiation, conflict resolution, exec reviews, dialogue scripts |
| M7 Analytics | Metrics definition, North Star, AARRR, funnels, A/B basics, AOV, SQL SELECT/WHERE, cohorts, CAC/LTV, contribution margin | Event instrumentation, GA4 demo | DAU/MAU stickiness, ARPU/ARPPU, payback, GMV, p-values, CIs, Type I/II, power & sample size, SQL joins/aggregation/window functions, Excel, Power BI, Mixpanel/Amplitude, experiment readouts |
| M8 Growth | GTM, acquisition/activation/monetisation, revenue models, case discussions | Pricing, referral | Growth loops, lifecycle marketing, PLG, freemium economics, upsell, churn reduction, launch retros, scaling, fintech/SaaS/consumer launch plans |
| M9 Tech 101 | Frontend/backend, APIs, HTTP methods/codes, SQL, CI/CD, webhooks, dev/staging/prod | Vector DBs | How the internet works, REST vs GraphQL, authN vs authZ, OAuth, caching, LB, queues, CDN, scaling, availability, security basics, system design (payments, food delivery, ride-hailing, notifications, onboarding) |
| M10 LLM / RAG | LLM basics, tokens, embeddings, RAG, fine-tuning vs RAG, prompting, AI PRDs, evals, ethics | Context windows | Transformer/attention (conceptual), vector DB options, hallucination mitigation, model selection, cost-latency trade-offs, LLM eval methods, monitoring, governance, RAG/inference diagrams |
| M11 Vibe Coding | Concept, Lovable/Replit/Claude/GitHub workflows, prototype ≠ production, security cautions | Cursor/Copilot/Bolt (named only in brief) | HTML/CSS/JS fundamentals, Git fundamentals, DB connection, deployment, testing generated code, end-to-end hands-on project |
| M12 Agentic AI | Agent definition, tool use, MCP vs API, orchestration, approval checkpoints, n8n, prompt-injection cautions | Memory, multi-agent (L49 doesn't teach it despite LMS text) | Agent architectures, memory systems, evaluation/observability, permission boundaries, failure recovery, cost/latency, agentic business models, financial-services support case |
Other brief requirements vs S1
| Requirement | S1 status | Action |
|---|---|---|
| 12 named real-company case studies | Only classroom exercises (YouTube Shorts, Blinkit AOV, Facebook Groups, Uber surge, Amazon RICE, Zomato persona, Claude teardown, dating) | Build 12 cases; where verified company data is unavailable, label as hypothetical scenario. No invented company results. |
| Framework library (29 frameworks) | ~10 appear | Build all 29 with visual, example, limits, template |
| 5 end-to-end projects | Capstone list only | Build 5 with datasets and model solutions |
| 100+ interview questions | Mock interview summaries (L22, L30, L52, L53); L44 missing | Build ≥100 with scoring rubric |
| Hands-on exercises | Light | Add per-topic exercises + answers |
Known weaknesses in S1 to correct (not reproduce)
Already flagged inside S1 itself — kept as "Source-flagged misconceptions": RICE (reach − impact ≠ churn; effort is the denominator); Ansoff quadrant labels; "all AI needs RAG/vector DB"; LoRA is a technique not a model; retrieval ≠ fine-tuning; vector DBs are not 3-D file stores; CORS ≠ all UI–API traffic; white-labeling ≠ dropshipping; sprint ≠ always two weeks; MVP ≠ prototype ≠ public launch; unicorn = $1B valuation, not revenue.
Additional observations made while reading S1:
- Dated product claims (pricing, plans, features of Figma Make, Lovable, Replit, n8n, Claude, ChatGPT, Gemini) go stale quickly → the handbook states capabilities generically and links to current official docs.
- S1 contains no worked statistics (p-values, CIs, power). M7 adds these with computed examples.
- S1 contains no SQL beyond SELECT … WHERE. M7/M9 add GROUP BY, JOIN, window functions on a provided sample dataset.
Provisional titles to verify against the screenshot
All 12 module titles, especially M11 "Vibe Coding for Product Managers" and M12 "Agentic AI and AI Systems" (likely truncated in the screenshot per the brief).