Designs end-to-end ML pipelines and deployment strategies for AI projects.
npx skills add https://github.com/jlaws/dotfiles --skill agent-ml-engineer-jlaws
Please help me install this Agent Skill. Skill: agent-ml-engineer Source: https://github.com/jlaws/dotfiles/tree/main/.agents/skills/agent-ml-engineer Command: npx skills add https://github.com/jlaws/dotfiles --skill agent-ml-engineer-jlaws
ML/AI architecture, training pipelines, deployment, and optimization are complex and error-prone. This skill provides structured guidance to design scalable, reproducible ML workflows and production-ready inference services.
Describe your ML project goals and constraints to receive an end-to-end pipeline plan.
High-intent search queries and answers about installing and using this skill.
Designing a machine learning pipeline requires structuring component selection, data preprocessing, model training, and experiment tracking. This skill provides structured guidance to orchestrate these training workflows, ensuring scalable and reproducible ML systems with clear evaluation criteria.
Orchestrating model training workflows involves constructing pipelines for data preprocessing, model training, and hyperparameter tuning. It ensures reproducible ML workflows by applying structured experiment tracking and clear guardrails throughout the training process.
Deploying machine learning models to inference services requires structured deployment and optimization strategies. This includes configuring monitoring, scaling, and A/B testing to ensure production-ready inference services that handle real-time requests efficiently.
Yes, this ML architecture approach applies to both production deployment and experimentation environments. It guides architecture decisions, component selection, and data flow design while ensuring scalable, reproducible ML workflows with clear evaluation criteria across any environment.
You should avoid unified ML pipeline architectures when your AI project lacks clear evaluation criteria or requires highly customized, non-reproducible workflows. Without structured guardrails and deployment strategies, complex ML pipelines become error-prone and difficult to scale effectively.
KBRglobalReview AI/ML architecture decisions and produce prioritized action plans.
bugrabilgeBuild production ML systems with PyTorch 2.x and TensorFlow.
Eigo-Mt-FujiGuide end-to-end ML model development and deployment workflows in Python with PyTorch or TensorFlow pipelines.
luiztrilhaDispatch subagents with isolated context for two-stage task reviews.
