nemotron-customize

Plan and execute Nemotron customization pipelines with per-step YAML configs.

Updated Apr 16, 2026

Search Tags:#planning #customization #organization #nemotron #pipelines #artifact-graph #yaml-config

Ready to run

npx skills add https://github.com/sayalinvidia/sayali-skills-test --skill nemotron-customize-sayalinvidia

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: nemotron-customize
Source: https://github.com/sayalinvidia/sayali-skills-test/tree/main/skills/nemotron-customize
Command: npx skills add https://github.com/sayalinvidia/sayali-skills-test --skill nemotron-customize-sayalinvidia

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This skill helps teams plan, configure, and execute end-to-end Nemotron customization workflows by stitching together curated data, model training steps, RL alignment, evaluation, and deployment-ready artifact graphs into cohesive pipelines.

Core Features & Use Cases

Orchestrates cross-step customization pipelines (curation, translation, SFT/PEFT, pretraining CPT, RL alignment, BYOB benchmarks, checkpoint conversion, and model optimization) using the official catalog, artifact graphs, and pattern guidance.
Generates per-step YAML configurations and ensures artifact compatibility across stages; routes user goals to the appropriate steps and validates prerequisites before execution.
Supports Explorer mode for plan-first workflows when no exact catalog surface exists, enabling safe scoping and planning before code changes.

Quick Start

Draft a complete Nemotron customization plan for a requested pipeline and generate a new config file under the target step's config/ directory.

Frequently Asked Questions about nemotron-customize

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I plan an end-to-end Nemotron customization pipeline across data curation and training?▼

To plan Nemotron customization, sequence curated JSONL translation, SFT/PEFT, pretraining CPT, RL alignment, and BYOB benchmarks into cohesive pipelines. The workflow routes goals through bundled catalogs, deduces artifact DAGs, and generates validated per-step YAML configs.

What is the process for generating YAML configs for Nemotron SFT and RL alignment steps?▼

Generating YAML configs for Nemotron SFT and RL alignment involves loading live step manifests from the catalog and producing new config files under src/nemotron/steps/<cat>/<step>/config/. The process ensures artifact compatibility and validates prerequisites before execution.

Can I use Explorer mode to scope a Nemotron customization plan before writing any config files?▼

Yes, Explorer mode supports plan-first workflows for Nemotron customization when no exact catalog surface exists. It enables safe scoping and planning of data curation, training, and evaluation steps before generating any actual YAML configuration files or code changes.

Does Nemotron customization support checkpoint conversion and model optimization workflows?▼

Yes, Nemotron customization pipelines support checkpoint conversion and model optimization. The workflow orchestrates these stages alongside pretraining CPT, RL alignment, and BYOB benchmarks, ensuring artifact compatibility across all stages using the official catalog and artifact graphs.

How do artifact graphs ensure compatibility across different stages of a Nemotron training pipeline?▼

Artifact graphs ensure compatibility across Nemotron training stages by deducing artifact DAGs that map dependencies between data curation, translation, SFT/PEFT, and RL alignment. The pipeline validates prerequisites against bundled ARTIFACTS and CATALOG references before generating step configs.

What are the limitations of using bundled catalogs for routing Nemotron customization tasks?▼

Nemotron customization relies on bundled ARTIFACTS, CATALOG, PATTERNS, and HARDWARE references to route tasks. If a requested pipeline lacks an exact catalog surface, users must switch to Explorer mode for safe scoping, as standard routing cannot deduce artifact DAGs for unmapped configurations.