Installs into .claude/skills of the current project.
Are you the author of Playbook Ml Research Scaling Decisions?
Add the live security badge to your README. It updates with every re-scan.
[](https://www.skillsdirectory.com/skills/aniruddhaadak80-playbook-ml-research-scaling-decisions)
---
name: playbook-ml-research-scaling-decisions
description: "Scaling-law claims converted into budget-honest training choices."
---
# Playbook: Ground scaling decisions in reality
> Scaling-law claims converted into budget-honest training choices.
**Track:** 🗺️ ML Research Engineering · **Domain:** Journey Playbooks · **Level:** journey · **~35 min**
**Who this is for:** Research Engineers, ML Scientists, PhD Researchers, Applied Scientists
## When to Use This Skill
Scaling-law claims converted into budget-honest training choices.
Use it whenever a matching task appears in conversation — the agent loads these instructions on demand.
## Journey Steps
1. Step 1 — Design ablations that isolate contributions: start with "List claimed components; rank by novelty and implementation cost"
2. How it fits together: Extrapolation beyond fitted compute ranges is strategy-by-vibes.
### Referenced Skills
- `experimentation-rigor-ablation-design`
## Commands
**Install all referenced skills**
```bash
npx skills add aniruddhaadak80/skills --skill experimentation-rigor-ablation-design
```
---
Part of [aniruddhaadak80/skills](https://github.com/aniruddhaadak80/skills) · Browse all at https://skills.sh/aniruddhaadak80/skills
