Skill UI

This deep dive guides users through migrating complex GPU workloads, such as deep learning training pipelines, from major hyperscalers (AWS, GCP, Azure) or other cloud platforms to Vast.ai. It covers critical steps including cost comparison analysis, adapting Docker images for new environments, securely handling cloud credentials via environment variables, and implementing robust validation and rollback procedures to ensure a smooth transition.

This guide details how to implement environment isolation (dev, staging, prod) for GPU cloud deployments on Vast.ai. It uses environment variables and configuration classes to enforce separate API keys, budget caps, instance limits, and GPU whitelists for each deployment tier, ensuring efficient resource management and cost control.

This comprehensive guide outlines security best practices for utilizing Vast.ai's cloud environment. It covers secure API key management, hardening SSH access to GPU instances, implementing proper data cleanup before instance destruction, and adhering to robust network security protocols. Use this when setting up, auditing, or deploying sensitive AI/ML workloads on shared cloud hardware.

Comprehensive guide for upgrading the Vast.ai CLI and Python SDK. This skill helps users manage API version migrations, detect breaking changes, and update associated Docker images (e.g., CUDA versions). It ensures continuous compatibility when working with cloud GPU computing resources.

Implement robust, event-driven workflows for managing GPU resources on Vast.ai. Since native webhooks are unavailable, this skill uses Python polling to continuously monitor instance status changes (e.g., loading, running, exited). It enables complex automation tasks such as auto-recovering from spot preemption, handling lifecycle events, and tracking billing costs.

A comprehensive guide covering modern web game development principles. It details framework selection (Phaser, Three.js), implementing WebGPU, performance optimization strategies like asset compression and object pooling, adhering to browser constraints, and utilizing PWA capabilities for cross-platform gaming experiences.

Detects system CPU, GPU, memory, and disk availability before heavy scientific tasks, writing `.claude_resources.json` with strategic recommendations (parallelism, out-of-core, GPU acceleration) so analysts choose appropriate compute approaches.

LatchBio Integration provides a Python-first platform to compose, deploy, and monitor serverless bioinformatics pipelines with the Latch SDK, decorator-based tasks, cloud storage (LatchFile/Dir), GPU tuning, and Nextflow or Snakemake interoperability.

PyTorch Lightning Toolkit streamlines PyTorch model training by organizing LightningModules, LightningDataModules, Trainer configs, callbacks, logging, and distributed strategies for scalable multi-GPU/TPU runs in research or production workflows.

A comprehensive skill for advanced text and typography handling within the Makepad framework. It provides robust mechanisms for font configuration, text layout management (Layouter), and high-quality GPU-based text rendering using Signed Distance Fields (SDF). Use this skill when needing to implement rich text displays, manage complex font styles, or ensure crisp, scalable typography across different resolutions.

This assistant specializes in the Makepad shader system. It helps users write, debug, and understand complex shader code using SDF2d and GLSL principles. Use it when developing custom widget backgrounds, advanced graphics, gradients, or visual effects that require low-level GPU rendering control within the Makepad framework.

Run NVIDIA Cosmos Policy evaluations on LIBERO and RoboCasa simulations, enabling headless EGL GPU rendering, inference profiling, and benchmark workflows via public eval entrypoints.