sagemaker-ai

Validates dataset formatting and quality for SageMaker model fine-tuning (SFT, DPO, or RLVR). Use when the user says "is my dataset okay", "evaluate my data", "check my training data", "I have my own data", or before starting any fine-tuning job. Detects file format, checks schema compliance against the selected model and technique, and reports whether the data is ready for training or evaluation.

Generates a Jupyter notebook that transforms datasets between ML schemas for model training or evaluation. Use when the user says "transform", "convert", "reformat", "change the format", or when a dataset's schema needs to change to match the target format — always use this skill for format changes rather than writing inline transformation code. Supports OpenAI chat, SageMaker SFT/DPO/RLVR, HuggingFace preference, Bedrock Nova, VERL, and custom JSONL formats from local files or S3.

Manages project directory setup and artifact organization. Use when starting a new project, resuming an existing one, or when a PLAN.md needs to be associated with a project directory. Creates the project folder structure (specs/, scripts/, notebooks/) and resolves project naming.

Selects a base model and fine-tuning technique (SFT, DPO, or RLVR) for the user's use case by querying SageMaker Hub. Use when the user asks which model or technique to use, wants to start fine-tuning, or mentions a model name or family (e.g., "Llama", "Mistral") — always activate even for known model names because the exact Hub model ID must be resolved. Queries available models, validates technique compatibility, and confirms selections.

Generates a Jupyter notebook that fine-tunes a base model using SageMaker serverless training jobs. Use when the user says "start training", "fine-tune my model", "I'm ready to train", or when the plan reaches the finetuning step. Supports SFT, DPO, and RLVR trainers, including RLVR Lambda reward function creation.

Generate comprehensive issue reports from HyperPod clusters (EKS and Slurm) by collecting diagnostic logs and configurations for troubleshooting and AWS Support cases. Use when users need to collect diagnostics from HyperPod cluster nodes, generate issue reports for AWS Support, investigate node failures or performance problems, document cluster state, or create diagnostic snapshots. Triggers on requests involving issue reports, diagnostic collection, support case preparation, or cluster troubleshooting that requires gathering logs and system information from multiple nodes.

Remote command execution and file transfer on SageMaker HyperPod cluster nodes via AWS Systems Manager (SSM). This is the primary interface for accessing HyperPod nodes — direct SSH is not available. Use when any skill, workflow, or user request needs to execute commands on cluster nodes, upload files to nodes, read/download files from nodes, run diagnostics, install packages, or perform any operation requiring shell access to HyperPod instances. Other HyperPod skills depend on this skill for all node-level operations.

Check and compare software component versions on SageMaker HyperPod cluster nodes - NVIDIA drivers, CUDA toolkit, cuDNN, NCCL, EFA, AWS OFI NCCL, GDRCopy, MPI, Neuron SDK (Trainium/Inferentia), Python, and PyTorch. Use when checking component versions, verifying CUDA/driver compatibility, detecting version mismatches across nodes, planning upgrades, documenting cluster configuration, or troubleshooting version-related issues on HyperPod. Triggers on requests about versions, compatibility, component checks, or upgrade planning for HyperPod clusters.

Generates a Jupyter notebook that deploys fine-tuned models from SageMaker Serverless Model Customization to SageMaker endpoints or Bedrock. Use when the user says "deploy my model", "create an endpoint", "make it available", or asks about deployment options. Identifies the correct deployment pathway (Nova vs OSS), generates deployment code, and handles endpoint configuration.

Generates a Jupyter notebook that evaluates a fine-tuned SageMaker model using LLM-as-a-Judge. Use when the user says "evaluate my model", "how did my model perform", "compare models", or after a training job completes. Supports built-in and custom evaluation metrics, evaluation dataset setup, and judge model selection.

Discovers user intent and generates a structured, step-by-step customization plan that orchestrates other skills. Always activate at the start of every conversation, when all tasks in a plan are completed, or when the user asks to modify the current plan. Handles intent discovery, plan generation, plan iteration, and mid-execution plan alterations. When in doubt, use this skill.

Creates a reusable use case specification file that defines the business problem, stakeholders, and measurable success criteria for model customization, as recommended by the AWS Responsible AI Lens. Use as the default first step in any model customization plan. Skip only if the user explicitly declines or already has a use case specification to reuse. Captures problem statement, primary users, and LLM-as-a-Judge success tenets.