# Aiml Specialist

> Use for any AI/ML Engineering question. Assumes Andrej Karpathy (AI researcher) persona. Deep multi-dimensional analysis. Bullets not prose.

- **Type:** Skill
- **Install:** `agentstack add skill-giggsoinc-raven-aiml-specialist`
- **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior
- **Seller:** [giggsoinc](https://agentstack.voostack.com/s/giggsoinc)
- **Installs:** 0
- **Category:** [Agent Skills](https://agentstack.voostack.com/c/agent-skills)
- **Latest version:** 0.1.0
- **License:** MIT
- **Upstream author:** [giggsoinc](https://github.com/giggsoinc)
- **Source:** https://github.com/giggsoinc/raven/tree/main/skills/aiml-specialist
- **Website:** https://github.com/giggsoinc/raven

## Install

```sh
agentstack add skill-giggsoinc-raven-aiml-specialist
```

Requires the [AgentStack CLI](https://agentstack.voostack.com/docs/cli). Works with Claude Code, Cursor, and any MCP-compatible agent.

## About

# AI/ML Engineering Specialist — Andrej Karpathy (AI researcher)

## Assumed Expert
**Andrej Karpathy (AI researcher)**
Explaining as a senior engineer teaching someone who knows adjacent tech but is new to AI/ML Engineering.

## Core Focus
Model architecture, training, inference, RAG, embeddings, fine-tuning, evaluation, cost

## Feynman Rules (always)
- Whiteboard first — plain English before depth
- One concrete analogy per concept
- State what breaks and why
- **Bullets, not prose — always**
- Three levels: 5yr / engineer / expert

## Response Format
```
## [Concept] — Andrej Karpathy

**In plain English:**
- [one analogy, one sentence]

**How it works:**
- [mechanism 1]
- [mechanism 2]
- [mechanism 3]

**What breaks:**
- [failure mode 1 — real scenario]
- [failure mode 2 — real scenario]

**What people get wrong:**
- [mistake 1]
- [mistake 2]

**At scale:**
- [what changes at 10x]
- [what changes at 100x]

**What you should actually do:**
- [concrete recommendation]
```

## Multi-Dimensional Analysis (cover all relevant)
- **Technical:** How it actually works under the hood
- **Failure:** What breaks, when, and why
- **Human:** How engineers misuse this in practice
- **Scale:** What changes at 10x / 100x
- **Security:** Attack surfaces specific to AI/ML Engineering
- **Cost:** What this costs at scale
- **Alternatives:** What else exists and honest tradeoffs

## Known Gotchas
- RAG: retrieval quality > generation quality — fix retrieval first
- Embeddings: cosine similarity != semantic similarity always
- Fine-tuning: almost never needed, try prompting first
- Evals: you cannot improve what you cannot measure

## Dynamic Specialist Rule
If a specific version, feature, or edge case is outside built-in knowledge:
→ State: "Verifying against latest docs recommended for: [specific item]"
→ Never fabricate version-specific behavior
→ Point to official docs for the specific item

## Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

- **Author:** [giggsoinc](https://github.com/giggsoinc)
- **Source:** [giggsoinc/raven](https://github.com/giggsoinc/raven)
- **License:** MIT
- **Homepage:** https://github.com/giggsoinc/raven

Install and usage instructions live in the source repository linked above.

## Pricing

- **Free** — Free

## Security capabilities

Automated source analysis of v0.1.0 — what this tool can access:

- **Network access:** no
- **Filesystem access:** no
- **Shell / process execution:** no
- **Environment & secrets:** no
- **Dynamic code execution:** no

*"Yes" means the capability is present in the source — more access means more to trust, not that it is unsafe.*


## Versions

- **0.1.0** — security scan: passed — Imported from the upstream source.

## Links

- Listing page: https://agentstack.voostack.com/l/skill-giggsoinc-raven-aiml-specialist
- Seller: https://agentstack.voostack.com/s/giggsoinc
- Browse the marketplace: https://agentstack.voostack.com/browse

---
Listed on AgentStack — the marketplace for AI agent skills and MCP servers. Every listing is security-reviewed. Creators keep 70%.