Skill UI

Automates guidance for configuring mlflow experiment tracking, covering best practices, code generation, and validation to support ML training workflows.

Automated guidance for managing model checkpoints in ML training workflows, offering best practices, sample code, and validation to keep experiments stable and reproducible.

Automates guidance on identifying and responding to model drift within ML deployment workflows, offering best practices, production-ready code, and validation support so you can keep inference systems stable.

Automated guidance for model explainability tooling within ML training workflows, giving step-by-step best practices, production-ready code, and validation for experiments whenever users mention explainability tool needs.

Guides ML deployment workflows by surfacing best practices, templated code, and validation steps whenever model pruning helper patterns are mentioned.

nanoGPT is a simplified, minimalist implementation designed specifically for educational purposes. It reproduces the core architecture of GPT-2 (124M) using clean, hackable code, allowing users to understand the entire transformer pipeline from scratch. The framework supports full workflows, including data preparation on various datasets (e.g., Shakespeare, OpenWebText), model training, and text generation, making it ideal for students and researchers learning NLP and deep learning principles.

Prediction Monitor Assistant automates machine learning deployment monitoring tasks, offering step-by-step guidance, best-practice patterns, production-ready code and configuration generation, and validation to keep serving and inference workflows reliable.

This skill provides implementation patterns and code examples for creating an incident runbook using Retell AI. It is specifically designed for automating complex workflows, handling emergency notifications, and managing interactive phone calls within a voice agent or telephony system environment.

Auto-activating assistant that guides AWS Sagemaker endpoint deployment, from best practices to production-ready code, configurations, and validation for inference serving and monitoring workflows.

Speeds up LLM inference by combining speculative decoding with Medusa multiple heads and lookahead techniques so chatbots, code assistants, and other real-time workloads can run with 1.5-3.6× lower latency while preserving target-model quality.

Automates TensorFlow SavedModel deployment support with step-by-step guidance, best-practice patterns, production-ready code/config generation, and validation for serving, monitoring, and optimization scenarios.

Automates guidance on train/test split workflows inside ML training, surfacing best-practice patterns, code snippets, and validation tips for PyTorch, TensorFlow, and sklearn projects.