Browse Skills

Analyzes GPU utilization and identifies performance bottlenecks during AI model training and inference tasks.

Simulates realistic dart throwing physics within DartSim environments, enabling training and analysis of robotic manipulation.

Generates clear, concise internal documentation and training materials based on provided context and desired style.

Delivers interactive, personalized cybersecurity training modules to employees, reducing phishing susceptibility and improving overall security posture.

This skill trains large language models like Megatron, accelerating AI development and enabling powerful custom LLM creation for diverse applications.

Provides LLMs guidance and assistance for building AI and machine learning applications.

This tool automatically adjusts gradients during training to prevent exploding gradients and improve model stability – a crucial aid for deep learning.

Analyzes user performance data to identify areas for improvement and suggest personalized training strategies.

Generates detailed, actionable skill descriptions and training data for custom AI agents based on expert knowledge.

Automates model checkpointing, versioning, and retrieval during training, simplifying experiment management and reproducibility.

This skill analyzes training effectiveness using Kirkpatrick’s levels – assessing if learning led to desired behavior changes and business results.

Automate Gemini model training checkpoints, versioning, and rollback for streamlined experimentation and reliable deployments.

Provides debugging guidance and assistance for software development workflows.

Automates Progressive Generation & Production Model deployment lifecycle management from training to serving.

This skill simulates Linux privilege escalation techniques for security testing and training, aiding in vulnerability identification.

Identifies and suggests fixes for Reinforcement Learning Model training errors based on logs and metrics.

Analyzes training logs to identify patterns and areas for improvement, boosting performance and efficiency through data-driven insights.

Leverages Redis caching to optimize AI training data access and reduce latency for teachingai models.

This skill automatically converts documents into vector datasets for efficient semantic search and AI model training.

This skill simulates Linux privilege escalation techniques for security testing and training, aiding in vulnerability identification.

Enables parallel experimentation using Git worktrees within LGBARN's backend for efficient model training and version control.

Simulates RL environments using UV Miles data for robust agent training and improved performance in realistic scenarios.

Uv Moe Training guides AI models to generate charming, cute (moe) aesthetics and styles for creative content generation.

This AI provides personalized workout plans and fitness advice, helping users achieve their health goals effectively.