How to Get Personalized Learning Recommendations with the find‑your‑level Skill

The find‑your‑level skill is an interactive ten‑question quiz that assesses your AI/ML knowledge and generates a customized study path through the AI Engineering from Scratch curriculum.

This skill eliminates guesswork when starting the rohitg00/ai-engineering-from-scratch curriculum by dynamically mapping your existing expertise to the correct entry point. By analyzing your responses to targeted multiple‑choice questions, the system produces personalized learning recommendations that tell you exactly which phases to skip, review, or study in depth.

How the find‑your‑level Skill Works

The skill operates as a conversational assessment tool embedded in AI agents such as Claude, Cursor, Codex, OpenClaw, or Hermes. When you invoke /find‑your‑level, the system executes a structured evaluation defined in [skills/find-your-level/SKILL.md](https://github.com/rohitg00/ai-engineering-from-scratch/blob/main/skills/find-your-level/SKILL.md).

Quiz Structure and Execution

The assessment presents ten multiple‑choice questions grouped into five knowledge‑area rounds. Each round tests a specific domain—such as Math & Statistics, Python Programming, or Machine Learning fundamentals—ensuring comprehensive coverage of prerequisite skills.

The quiz definition and question logic reside in the skill manifest, while the authoritative answer key lives in [skills/find-your-level/references/answer-key.md](https://github.com/rohitg00/ai-engineering-from-scratch/blob/main/skills/find-your-level/references/answer-key.md). This separation allows the skill to grade responses against canonical solutions while keeping the interaction logic modular.

Scoring Logic and Entry Point Mapping

Each question carries a value of one point, producing a total score between 0 and 10. After each round, the skill tallies your area‑specific performance, and after the final round, it applies a Score‑to‑Entry‑Point mapping (defined in lines 66‑74 of SKILL.md) to determine your optimal starting position within the 20‑phase curriculum.

For example, a score of 7/10 typically maps to Phase 11 – LLM Engineering, bypassing foundational topics you have already mastered.

Generating Your Personalized Learning Path

Once the entry point is identified, the skill constructs a full curriculum roadmap tailored to your assessment results.

Dynamic Curriculum Table Construction

The skill generates a markdown table listing all 20 phases of the curriculum, assigning each a status of Skip, Review, or Do based on your performance. Critically, the skill reads phase hour estimates dynamically from the canonical [ROADMAP.md](https://github.com/rohitg00/ai-engineering-from-scratch/blob/main/ROADMAP.md) (as referenced in lines 91‑95 of the manifest) to ensure time projections remain current with curriculum updates.

This dynamic lookup means your personalized plan always reflects the latest content estimates without manual recalculation.

Persistence and Next Steps

The final output presents a concise recommendation summarizing total estimated hours and specific phase priorities. To operationalize your personalized learning recommendations, the skill prompts you to run /start‑learning, which persists the placement into a LEARNING.md study plan file. Once saved, the /learn command launches the first lesson interactively, beginning with your designated entry phase.

Using the Skill in Practice

Invoke the skill through any compatible agent interface. Below is a representative interaction showing the quiz format and resulting roadmap:

User: /find-your-level
Assistant:
> **Round 1 – Math & Statistics**
> Q1: What is the dot product of a=[1,2,3] and b=[4,5,6]?
> A) 32  B) 21  C) 15  D) 27  
> Q2: Probability of exactly 2 heads in 3 fair‑coin flips?
> A) 1/4  B) 1/2  C) 1/8  D) 3/8  
> Reply with `Q1: <letter>, Q2: <letter>.`

[... rounds 2‑5 ...]

**Score:** 7/10  
**Entry Point:** Phase 11 – LLM Engineering  

| Phase | Name                | Status | Est. Hours |
|-------|---------------------|--------|------------|
| 0     | Setup & Tooling     | Skip   | --         |
| 1     | Math Foundations    | Skip   | --         |
| ...   | ...                 | ...    | ...        |
| 11    | LLM Engineering     | Do     | 25         |
| 12    | Prompt Engineering  | Review | 12         |
| ...   | ...                 | ...    | ...        |

Your personalized path: ~37 hours across 8 phases.  
**Recommendation:** Begin with Phase 11 (LLM Engineering) and focus on the "Prompt Engineering" review topics.  

*To save the plan:* `/start-learning`  
*To start the first lesson:* `/learn`

Summary

• The find‑your‑level skill in rohitg00/ai-engineering-from-scratch provides data‑driven personalized learning recommendations through a ten‑question diagnostic quiz.
• Scoring logic in skills/find-your-level/SKILL.md maps your 0‑10 score to a specific curriculum entry point (Phase 0‑19).
• The skill dynamically pulls hour estimates from ROADMAP.md to generate an accurate, up‑to‑date study plan with Skip, Review, and Do statuses.
• Use /start‑learning to persist your plan to LEARNING.md, then /learn to begin interactive lessons at your designated level.

Frequently Asked Questions

How do I invoke the find‑your‑level skill?

You can trigger the skill by typing /find‑your‑level in any AI agent that has the curriculum skills installed, including Claude, Cursor, Codex, OpenClaw, or Hermes. The skill manifest is located at .claude/skills/find-your-level/SKILL.md for Claude‑Code agents and skills/find-your-level/SKILL.md in the public repository.

What determines my entry point in the curriculum?

Your entry point is determined by the Score‑to‑Entry‑Point mapping defined in lines 66‑74 of the skill manifest. After calculating your raw score (0‑10) against the answer key in references/answer-key.md, the skill selects the phase where your knowledge gaps begin, allowing you to skip foundational material you have already mastered.

Can I change my learning plan after completing the quiz?

Yes. After receiving your personalized roadmap, execute /start‑learning to save the placement to a LEARNING.md file. If you need to reassess later, simply run /find‑your‑level again to generate a new recommendation based on updated knowledge, then overwrite your existing plan with another /start‑learning command.

Where does the skill get the time estimates for each phase?

The skill reads hour estimates dynamically from the canonical ROADMAP.md file (as specified in lines 91‑95 of SKILL.md). This ensures that your personalized schedule reflects the most current time projections whenever the curriculum maintainers update phase durations.

{
  "mcpServers": {
    "instagit": {
      "command": "npx",
      "args": ["-y", "instagit@latest"]
    }
  }
}