Explore our collection of Agent Skills to enhance your AI workflow.
Monitors GPU performance, VRAM allocation, and inference metrics for Ollama models to optimize AI workloads.
Orchestrates complex build workflows for OS images, container pods, and VM artifacts with intelligent branch-based caching.
Facilitates direct REST API operations for local Ollama instances using the Python requests library.
Automates the complete Git workflow by managing branches, committing changes, and streamlining GitLab merge requests.
Manages visual settings, GTK themes, and terminal aesthetics for the Bazzite operating system.
Manages disk partitions, automounting, and BTRFS filesystem features for Bazzite Linux environments.
Enables seamless integration between Ollama and the OpenAI ecosystem using a compatible API layer for popular libraries.
Manages bootable container virtual machines using bcvk for rapid testing and disk image exporting.
Simplifies the installation of standalone development tools, AI utilities, and Flatpak applications on Bazzite systems.
Accelerates LLM inference by up to 2x using Unsloth and vLLM backends for optimized model execution.
Configures LUKS disk encryption, TPM auto-unlock, and system-level security settings for Bazzite Linux.
Manages LocalAI inference services with GPU acceleration using Podman Quadlet containers for local model hosting.
Imports GGUF models directly from HuggingFace into Ollama for immediate local inference and testing.
Implements Retrieval-Augmented Generation workflows to ground AI responses with external knowledge and private documents.
Manages and configures Open WebUI instances for interacting with Ollama models via Podman containers.
Manages multiple JupyterLab development environments with GPU acceleration using Podman Quadlet containers.
Manages Apptainer containers for HPC workloads with automatic GPU passthrough and SIF image handling.
Automates the end-to-end Git workflow by committing changes, pushing branches, and creating pull requests with Japanese descriptions.
Provides expert guidance and implementation patterns for fine-tuning Large Language Models using PyTorch and HuggingFace.
Optimizes LLM alignment through Group Relative Policy Optimization (GRPO) for stable reinforcement learning and reasoning model training.
Maps and explains the codebase structure, module responsibilities, and technology stack of the LomoMarketplace monorepo.
Standardizes TypeScript code quality through robust error handling patterns, custom error classes, and resilient retry logic.
Generates and refines DuckDB SQL queries for data analysis using local database, CSV, and Parquet files.
Fine-tunes and optimizes vision-language models like Pixtral and Ministral using Unsloth's FastVisionModel and LoRA.
Manages and monitors aggregate AI service containers and pods within the Bazzite ecosystem.
Trains reward models for RLHF pipelines using preference datasets and sequence classification heads.
Manages overlay testing sessions and live editing of justfiles on immutable Linux systems for Bazzite-AI development.
Conducts autonomous risk assessments and root cause analysis during design phases to ensure robust architectural decisions.
Verifies Bazzite AI installations through comprehensive runtime tests for GPU detection, CUDA functionality, and service health.
Streamlines UI layout and positioning in Unreal Engine using standardized ReactUMG slot mapping rules.
Scroll for more results...