Career Frameworks

A 3-step career planning framework suite: build your professional narrative, read where real power lives in any company, and decide which roles to chase. AI-assisted, goal-agnostic.

Created by Jonathan Oh — originally a personal toolkit, adapted here so anyone can use it.

---

What this is

Three frameworks that work together as a system:

1. Master Narrative — your single source of truth, compressed into resume / LinkedIn / personal site / recruiter pitch variants.
2. Core Function Alignment — a diagnostic for reading where real power lives inside any company, so titles and prestige stop fooling you.
3. Next Role Strategy — a decision system that ranks role types against your specific goal and tells you what to chase, what to evaluate, and what to walk away from.

Each framework comes as two files: a template (the shape of the output) and an intake guide (the questionnaire + AI prompt that fills the template in for you).

This is not a resume builder. It's not a job board. It's not a personality assessment. It's a thinking tool designed for people who are about to make a real career decision and want to do it with a doctrine, not vibes.

Quick start

1. Open README.md (you're here). Read it.
2. Open 01_master_narrative_intake.md. Answer the 30 numbered questions.
3. Copy the prompt block at the bottom of that file into Claude or ChatGPT,
   paste your answers underneath, hit send. You'll get a filled-in
   Master Narrative back. Save it.
4. Repeat for Steps 2 and 3.

Time budget: ~5-8 hours total across all 3 steps. Can be split across sessions.

File inventory

File	What it's for
README.md	This file. Start here.
01_master_narrative_template.md	Blank shape of the Master Narrative output.
01_master_narrative_intake.md	30-question intake + AI prompt for Step 1.
02_core_function_alignment_template.md	Blank shape of the Core Function Alignment output.
02_core_function_alignment_intake.md	16-question intake + AI prompt for Step 2.
03_next_role_strategy_template.md	Blank shape of the Next Role Strategy output.
03_next_role_strategy_intake.md	32-question intake + AI prompt for Step 3.
examples/	Filled-in examples for fictional personas — "what good looks like" before you start.
CONTRIBUTING.md	How to suggest improvements (open an issue first).
LICENSE	MIT. Use it however you want.

The 3-step suite

Step 1: Master Narrative   →   Step 2: Core Function   →   Step 3: Next Role
─────────────────────          ────────────────────         ─────────────────
Who am I and what's            Where would I have           Which opportunities
my evidence?                   real power and learn         actually close my
                               the most?                    gaps fastest?

The sequence matters:

• Step 1 → Step 2: Step 2 references your evidence stack from Step 1 when assessing where you'd have leverage.
• Step 2 → Step 3: Step 3 uses Step 2's archetype lens as a filter inside its tier system.

You can do them out of order, but you'll be working harder than you have to.

Step 1 — Master Narrative

Builds your single source of truth. One document that captures your projects, work history, education, skills, and positioning — then compresses each piece into format-specific variants (resume, LinkedIn, personal site, recruiter pitch).

The point: stop rewriting your story for every audience. Write it once with the right structure, then tailor.

Step 2 — Core Function Alignment

Teaches you to read where real power lives inside any company. Five archetypes (research-led, product-led, engineer-led, sales-led, deployment-led), a power equation, and a tier system that classifies any role from Core to Support.

The point: stop being fooled by titles and prestige. Find out whether a role would actually let you do the thing you say you want to do.

Step 3 — Next Role Strategy

Turns your goal into a decision system. Define a north-star goal, identify the capability gaps between where you are and where you need to be, then rank role types into Tier 1/2/3 based on how directly each closes those gaps. Plus diligence questions, recruiter talking points, and a compensation philosophy.

The point: stop evaluating opportunities one at a time. Evaluate them against a doctrine you wrote when you were thinking clearly.

What makes this different from generic career advice

Three things, all baked into the methodology:

1. Audience-routing (Master Narrative): every piece of content has explicit "use this framing for X, not for Y" guidance. You stop discovering the right framing inside the next live conversation.
2. Archetype-first classification (Core Function Alignment): you classify the company before you evaluate the role. Same job title means radically different things across archetypes — a PM at Figma is not the same as a PM at Stripe.
3. Gap-driven evaluation (Next Role Strategy): roles are evaluated against the specific capability gaps between you and your goal, not against generic "is this a good job?" criteria.

If those three ideas feel like overkill for your situation, this probably isn't the right framework for you. If they feel like exactly what you've been missing, keep going.

How to use AI-assisted mode

Each intake guide has three parts:

• Part A — Methodology primer: ~200 words explaining what the framework does and why this structure exists.
• Part B — Structured questionnaire: numbered questions you answer in writing. This is the slow part — the AI cannot make up your life for you.
• Part C — AI prompt block: a copy-paste prompt that consumes your answers and produces a filled-in template.

The flow:

1. Read Part A so you understand what you're building.
2. Answer Part B in a text editor. Be specific. Vague answers produce vague output.
3. Copy the entire Part C prompt block (everything inside the triple backticks).
4. Paste it into Claude / ChatGPT / Gemini, then paste your numbered answers underneath.
5. The AI produces your filled-in framework. Save it as [yourname]_master_narrative_v1.md (or similar).
6. Read it critically. Edit anywhere it feels false, generic, or self-aggrandizing.
7. Move to the next step.

The prompts include instructions for the AI to flag thin sections rather than invent content. When you see ⚠️ Insufficient input — please answer: [X], go back and add more.

For AI agents helping a user

This section is for AI assistants. If you're a human reading along, you can skip it.

If you're an AI assistant (Claude, ChatGPT, Gemini, or similar) and a user has shared this repo with you and asked for help, here's the operating model:

The user is the source of truth about their own career. Never invent projects, metrics, dates, outcomes, companies, or accomplishments. If their input is too thin to fill a section, flag it as ⚠️ Insufficient input — please answer: [specific follow-up question] rather than improvising.

Which step are they on?

• If they haven't picked a step, point them at 01_master_narrative_intake.md and ask them to complete it.
• If they say "Step 1" or "Master Narrative," consume 01_master_narrative_intake.md (Part C contains the canonical prompt — execute it).
• If they say "Step 2" or "Core Function," consume 02_core_function_alignment_intake.md.
• If they say "Step 3" or "Next Role Strategy," consume 03_next_role_strategy_intake.md.

The intake guides contain the canonical prompts. Part C of each intake guide IS the prompt you should execute. Don't paraphrase it — use it. The prompt encodes the methodology constraints and the exact output format.

Methodology preservation is non-negotiable. These frameworks have three distinctive features that must NOT be diluted:

• Audience-routing (Step 1): every piece of content gets explicit "when to use" / "when not to use" guidance per audience type.
• Archetype-first classification (Step 2): always classify the company into one of five archetypes BEFORE evaluating any role.
• Gap-driven evaluation (Step 3): roles are evaluated against the user's specific gaps, not against generic criteria.

If you find yourself producing output that looks like generic career advice, stop and re-anchor to the methodology.

Format-variant compression (Step 1 only): every project, role, and skill gets at least three format variants — Resume (tight bullets, outcome-first, numbers), LinkedIn (narrative, 2-4 sentences, business-impact-forward), Website (full story, philosophy, technical detail). Don't collapse them.

The 60/40 principle (Steps 2 and 3): every role should be ~60% leveraging existing strengths, ~40% uncomfortable growth. Apply this check after tier classification.

The kill-switch question (Steps 2 and 3): "Am I the engine or the wheels?" If a role would put the user in a support function relative to the core group, flag it as a high-bar decision regardless of brand or comp.

Goal-agnostic framing: the original frameworks were written for a founder-track goal. The published version generalizes to any goal (exec leadership, IC craft mastery, career pivot, financial independence, etc.). Honor the user's stated north-star — don't substitute the original's founder-track assumptions.

If asked to skip the questionnaire and "just generate something": push back. Suggest the user answer at least the high-leverage questions first (Step 1: identity, current role, top 2 projects, 1-2 audience types; Step 2: north-star goal, 3-5 gaps, 1-2 target companies; Step 3: goal, gaps, top 2-3 role types under consideration). Without input, the output is generic by definition.

Adapting to your situation

The frameworks are written to be adaptable:

Not technical? Step 1's "Section 6: Engineering Depth" can be renamed "Craft Depth," "Domain Depth," "Operating Depth," or whatever fits your craft. The pattern (evidence-stack-ranked, methodology, daily workflow, framing guidance) is universal; the label adapts.

Not founder-track? Step 3's north-star can be exec leadership, IC craft mastery, career pivot into a new field, financial independence, transitioning out of a specific career, or anything else with a 2-7 year horizon. The Tier 1/2/3 system and kill-switch questions work regardless of goal.

Solo / non-traditional career? The "Professional Experience" sections accommodate self-employment, contract work, sabbaticals, and parenting / caregiving breaks. Frame each chapter the same way: problem → what you did → outcome → format-specific variants.

Confidential work? The Master Narrative template has explicit confidentiality guideline sections per project. Use them. The framework was originally built by someone whose work is heavily NDA-bound; this is a first-class concern, not an afterthought.

How long this takes

• Step 1: 2-4 hours of intake (you're inventorying your career). Most of that is recalling details about past projects.
• Step 2: 30-60 minutes (lighter intake — it's mostly conceptual).
• Step 3: 1-2 hours, plus ongoing updates as you have conversations with recruiters and companies.

You don't have to do this in one sitting. The intake guides are designed to be picked up and put down.

A note on honesty

The output is only as good as the inputs. If you describe a project in vague metaphors, you'll get vague positioning. If you describe outcomes with specific numbers, audiences, and decisions you owned, you'll get sharp positioning. The intake guides ask for specifics on purpose. Take the time to answer them properly.

The frameworks include explicit "accuracy guardrails" sections that ask you to write down things you might be tempted to overclaim under interview pressure. Use them. The version of yourself that's calm and writing in a text editor is more honest than the version that's three sentences into a panel interview.

Contributing

See CONTRIBUTING.md. Short version: open an issue before a PR. Improvements should sharpen the distinctive methodology, not dilute it.

License

MIT. Use it for yourself, your friends, your team. Fork it, modify it, build on top of it. Attribution appreciated but not required.

---

Created by Jonathan Oh. Built originally as a personal toolkit while figuring out my own next move. Open-sourced because friends kept asking for it.

If you've used these frameworks and they helped, I'd love to hear about it. If they didn't, I'd love to hear that too.