Skill UI

Ray Data offers scalable distributed data processing for ML workloads, streaming across CPU/GPU clusters and integrating with PyTorch, TensorFlow, and Ray Train. Use it for batch inference, ETL, and multi-modal preprocessing from a laptop up to hundreds of nodes.

This skill proactively detects and reports comprehensive system resources, including CPU cores, GPU types (NVIDIA, AMD, Apple Silicon), available memory, and disk space. It generates a detailed JSON report with strategic, context-aware recommendations for optimizing scientific computing workflows, such as recommending parallel processing libraries (Dask, joblib), out-of-core methods (Zarr), or appropriate GPU acceleration backends (Metal, ROCm). Use this before any large-scale data analysis, model training, or computationally intensive simulation to make informed architectural decisions.

Latch is a robust Python framework designed for building, deploying, and managing complex, scalable bioinformatics workflows as serverless pipelines. It supports integrating multiple analysis tools (like Nextflow/Snakemake) and handles cloud data management (LatchFile/LatchDir) and detailed resource configuration (GPU, CPU, memory). Ideal for running large-scale analyses such as RNA-seq, AlphaFold, and single-cell analysis in a reproducible, production-ready environment.

This skill guides users through building, deploying, and maintaining machine learning applications on Hugging Face Spaces. It provides comprehensive knowledge on selecting the optimal SDK (Gradio, Docker, Static), choosing appropriate hardware resources (ZeroGPU, dedicated GPUs, CPU-basic), and following best practices to ensure ML applications are deployed efficiently and successfully on the platform.