Skill UI

Automates dataset loader creation within ML Training workflows, offering best-practice guidance, validation checks, and production-ready code to accelerate data prep and model tuning.

Automated guidance for hyperparameter tuning within ML training workflows, offering best practices, production-ready code, and validation across modeling stages like data prep, training, and experiment tracking.

This skill provides comprehensive guidance for building production-grade MLOps pipelines. It covers the entire machine learning lifecycle, from initial data ingestion and rigorous feature engineering to model training, advanced validation (A/B testing, performance drift), and robust deployment strategies (Canary, Blue-green). It enables the design and orchestration of scalable, reproducible, and observable ML systems using tools like Airflow and Kubeflow.

A comprehensive checklist for deploying large-scale, production-ready GPU workloads on Vast.ai. It covers critical lifecycle stages including account authentication, instance validation, data encryption, persistent checkpointing strategies, fault tolerance for spot instances, robust monitoring (utilization & cost), and strict cost controls, ensuring reliable and stable ML/AI deployments.

Automates wandb experiment logger tasks within ML training, providing guidance, best practices, and code or configuration examples plus validation to keep data prep, training, hyperparameter tuning, and experiment tracking consistent.

PydanticAI is a robust Python framework that integrates Pydantic's type-checking and validation capabilities into Large Language Models (LLMs). It allows developers to build production-ready AI agents that guarantee structured, validated outputs, moving beyond simple raw strings. Key features include supported tool calling, dependency injection for testability, and multi-model compatibility across major providers like OpenAI, Anthropic, and Gemini. Use this when reliability and correct data structure are paramount.

This framework implements robust input and output validation guardrails for LLM-powered applications. It prevents prompt injection, detects and redacts PII, filters toxic content, and validates model outputs against defined schemas. Use it when deploying AI agents that handle sensitive data or require strict adherence to corporate policies, ensuring a secure and trustworthy AI interaction layer.

This comprehensive skill provides multi-layered guardrails for integrating with LLMs like Claude. It enforces security and compliance by implementing input validation (PII detection, length checks), system prompt restriction (scope control, safety rules), output filtering (preventing sensitive data leakage), cost governance, and model access policies. Ideal for building secure, robust, and governed AI applications.

This skill guides the secure handling and lifecycle management of large-scale data artifacts, including training datasets, model weights, and checkpoint snapshots, within GPU cloud environments. It provides best practices for data pipelines, covering secure import via Kubernetes PVCs, compliant export, and stringent data validation, ensuring adherence to encryption standards (AES-256) and regulatory requirements (PII masking, RBAC).

This skill utilizes NVIDIA's open-source garak framework to conduct comprehensive red-teaming assessments on Large Language Models (LLMs). It tests for critical vulnerabilities such as prompt injection, jailbreaks, data leakage, and toxic content generation by sending thousands of adversarial probes. Ideal for pre-deployment security validation, API guardrail testing, and generating defensible evidence for AI risk assessments.

This skill provides comprehensive monitoring advisory services for data assets and AI agents. It helps users assess data coverage gaps, analyze use-case criticality, and automatically create data quality monitors (table, metric, validation). Furthermore, it enables monitoring of AI agent performance, latency, and observability, ensuring robust data governance across the entire ML lifecycle.