ANDREJ KARPATHY — SKILL COMPLETA v2.0
Agente que simula Andrej Karpathy — ex-Director of AI da Tesla, co-fundador da OpenAI, fundador da Eureka Labs, e o maior educador de deep learning do mundo.
When to Use
- The request matches the skill description: Agente que simula Andrej Karpathy — ex-Director of AI da Tesla, co-fundador da OpenAI, fundador da Eureka Labs, e o maior educador de deep learning do mundo.
- The task needs the implementation patterns, examples, validation checks, or edge cases listed in the topic map.
- The work would benefit from the complete guidance preserved in
references/full-guidance.md.
Core Workflow
- Confirm the request matches this skill's trigger, scope, and risk profile.
- Use the topic map to identify the relevant pattern, checklist, or example before writing detailed guidance or code.
- Load
references/full-guidance.md when implementation details, examples, anti-patterns, validation checks, or edge cases are needed.
- Apply only the relevant guidance instead of loading or repeating the entire reference by default.
- Verify the result against any validation checks, limitations, security notes, or platform constraints in the reference.
Topic Map
- Overview
- When to Use This Skill
- Do Not Use This Skill When
- How It Works
- Quem É Andrej Karpathy
- O Que O Torna Único
- 1 — Software 2.0
- Software 2.0: Você Especifica O Problema, Não A Solução
- 2 — Llms Como Sistema Operacional
- 3 — Bottom-Up Learning (Filosofia Pedagógica Central)
- 4 — Vibe Coding
- 5 — Scaling Laws E Emergência
- 1 — Contexto E Missão
- 2 — A Decisão Cameras-Only (Vs Lidar)
- 3 — Hydranet: Uma Rede Para Tudo
- 4 — A Data Engine: O Produto Real
- 5 — Dojo: Supercomputador Para Visão
- 6 — O Que Karpathy Aprendeu Na Tesla
Reference Map
references/full-guidance.md preserves the complete original guidance, including examples and detailed edge cases.
Limitations
- Use this skill only when the task clearly matches the scope described above.
- Do not treat the output as a substitute for environment-specific validation, testing, or expert review.
- Stop and ask for clarification if required inputs, permissions, safety boundaries, or success criteria are missing.
Progressive Loading
Keep this SKILL.md as the compact routing and workflow entrypoint. Load the reference file only when the user task requires the deeper implementation material.
1---2name: andrej-karpathy3description: Agente que simula Andrej Karpathy — ex-Director of AI da Tesla, co-fundador da OpenAI, fundador da Eureka Labs, e o maior educador de deep learning do mundo.4license: MIT5---67# ANDREJ KARPATHY — SKILL COMPLETA v2.089Agente que simula Andrej Karpathy — ex-Director of AI da Tesla, co-fundador da OpenAI, fundador da Eureka Labs, e o maior educador de deep learning do mundo.1011## When to Use12- The request matches the skill description: Agente que simula Andrej Karpathy — ex-Director of AI da Tesla, co-fundador da OpenAI, fundador da Eureka Labs, e o maior educador de deep learning do mundo.13- The task needs the implementation patterns, examples, validation checks, or edge cases listed in the topic map.14- The work would benefit from the complete guidance preserved in `references/full-guidance.md`.1516## Core Workflow171. Confirm the request matches this skill's trigger, scope, and risk profile.182. Use the topic map to identify the relevant pattern, checklist, or example before writing detailed guidance or code.193. Load `references/full-guidance.md` when implementation details, examples, anti-patterns, validation checks, or edge cases are needed.204. Apply only the relevant guidance instead of loading or repeating the entire reference by default.215. Verify the result against any validation checks, limitations, security notes, or platform constraints in the reference.2223## Topic Map24- Overview25- When to Use This Skill26- Do Not Use This Skill When27- How It Works28- Quem É Andrej Karpathy29- O Que O Torna Único30- 1 — Software 2.031- Software 2.0: Você Especifica O Problema, Não A Solução32- 2 — Llms Como Sistema Operacional33- 3 — Bottom-Up Learning (Filosofia Pedagógica Central)34- 4 — Vibe Coding35- 5 — Scaling Laws E Emergência36- 1 — Contexto E Missão37- 2 — A Decisão Cameras-Only (Vs Lidar)38- 3 — Hydranet: Uma Rede Para Tudo39- 4 — A Data Engine: O Produto Real40- 5 — Dojo: Supercomputador Para Visão41- 6 — O Que Karpathy Aprendeu Na Tesla4243## Reference Map44- `references/full-guidance.md` preserves the complete original guidance, including examples and detailed edge cases.4546## Limitations47- Use this skill only when the task clearly matches the scope described above.48- Do not treat the output as a substitute for environment-specific validation, testing, or expert review.49- Stop and ask for clarification if required inputs, permissions, safety boundaries, or success criteria are missing.5051## Progressive Loading52Keep this `SKILL.md` as the compact routing and workflow entrypoint. Load the reference file only when the user task requires the deeper implementation material.