Brainstorm ML/AI decisions and recommend paths from repository context.
npx skills add https://github.com/jayll1303/AIEKit --skill ml-brainstorm
Please help me install this Agent Skill. Skill: ml-brainstorm Source: https://github.com/jayll1303/AIEKit/tree/main/.kiro/skills/ml-brainstorm Command: npx skills add https://github.com/jayll1303/AIEKit --skill ml-brainstorm
Brainstorm ML/AI decisions and recommended paths given project context and constraints.
Evaluate training strategy options (Full fine-tune, LoRA, QLoRA, Unsloth) for a given context and output a recommended plan with linked skills.
High-intent search queries and answers about installing and using this skill.
To decide on an ML training strategy, evaluate approaches like full fine-tune, LoRA, QLoRA, or Unsloth by scanning your repository context and constraints. The skill generates viable options with pros/cons and a recommended path.
The best way to select a serving engine and quantization method is to brainstorm ML decisions using your project context. This evaluates multiple approaches, factoring in VRAM considerations, and outputs a structured pipeline architecture plan.
Yes, you can plan pipeline architecture for diverse ML codebases by scanning repository context. It evaluates multiple approaches and outputs a plan with pros/cons, VRAM considerations, and recommended skill chains.
To evaluate model selection options under project constraints, apply context-aware decision brainstorming. It generates viable approaches with pros/cons and VRAM considerations to recommend a tailored path.
Generating an ML architecture plan outputs a structured skill chain linking to AIEKit workflows and reference materials. This provides a recommended path connecting your brainstormed decisions to actionable workflows.
fcastrilloDetect project type, languages, and key directories into work/discovery/context.yaml.
KaiserWhoLearnsValidate ML data, models, and pipelines with structured reports.
ascend-ai-codingServe OpenAI-compatible LLM inference on Huawei Ascend NPUs.
nahisahoDevelop end-to-end ML models with training, evaluation, deployment, and MLOps.
