Skill UI

A specialized skill for writing highly efficient, memory-safe, and robust C code. This guide focuses on best practices for systems programming, including rigorous memory management (malloc/free), advanced pointer arithmetic, POSIX compliance, multi-threading (pthreads), and embedded systems constraints. It emphasizes using static analysis tools, implementing comprehensive error handling, and ensuring performance through proper profiling. Ideal for complex, low-level system tasks.

This guide details the use of the GGUF format and quantization techniques for running large language models (LLMs) efficiently. GGUF enables optimal inference on consumer hardware, Apple Silicon, and systems requiring CPU-only operation. By compressing models using various K-quant methods (like Q4_K_M), developers can significantly reduce memory footprint and hardware requirements while maintaining high performance.

This guide provides comprehensive techniques to optimize LangChain applications for production environments. Learn how to measure baseline latency, implement efficient caching strategies (in-memory, Redis, SQLite), utilize batch processing with concurrency control, and maximize perceived speed through streaming. Furthermore, it covers prompt optimization and strategic model selection to ensure high throughput and low costs.

llama.cpp is a pure C/C++ library designed for efficient LLM inference, optimized to run on resource-constrained environments. It excels in scenarios where high-end NVIDIA GPUs (CUDA) are unavailable, making it ideal for Apple Silicon Macs, AMD/Intel GPUs, or edge deployment devices like Raspberry Pi. It supports GGUF quantization formats, enabling significant memory reduction and speed improvements for local, cross-platform AI applications.

This comprehensive guide provides step-by-step instructions to optimize localization spending on Lokalise. Learn how to audit current usage across projects and languages, minimize per-seat costs by managing contributor access, maximize Translation Memory (TM) hits, and efficiently triage Machine Translation (MT) to keep your localization projects well within budget.

An expert guide for performing complex data analysis, cleaning, and transformation using pandas DataFrames. This skill covers advanced techniques such as vectorized operations, efficient merging, time series resampling, group-by aggregations, and memory optimization, ensuring production-grade code and robust data validation.

Quantizes HuggingFace LLMs to 8-bit or 4-bit with bitsandbytes, cutting memory by up to 75% while keeping accuracy, and supports QLoRA fine-tuning plus 8-bit optimizers for faster, memory-efficient training.

This guide provides advanced patterns and best practices for optimizing Apache Spark jobs. It covers critical areas like effective partitioning, memory tuning, minimizing data skew, and optimizing different types of joins (broadcast, bucket). Use this when troubleshooting slow, large-scale data processing workflows or needing to scale Spark pipelines efficiently and reliably.

This comprehensive guide offers detailed steps to optimize the Windsurf IDE experience. It addresses common performance bottlenecks such as slow indexing, high memory consumption, and inefficient AI context building (Cascade). Users will learn best practices for configuring exclusion patterns, managing focused workspaces, and tuning IDE settings for maximum development efficiency.

A specialized review tool for identifying critical performance bottlenecks within Django applications. It focuses on validating issues such as N+1 queries, unbounded querysets, missing database indexes, and write loops. Use this when you suspect database slowdowns or memory exhaustion due to inefficient ORM usage, ensuring findings are provable and actionable.

This skill provides a comprehensive guide to implementing high-performance, type-safe, in-memory caching in Go. It covers advanced features including nine eviction algorithms (e.g., W-TinyLFU, LRU), Time-To-Live (TTL), read-through loaders, and robust capacity sizing techniques. It is ideal for backend services requiring reliable, low-latency data access and efficient resource management.

Provides programmatic access to massive, versioned public single-cell and spatial transcriptomics datasets from CZ CELLxGENE. This skill allows users to efficiently query specific subsets of cell metadata, gene expression matrices, embeddings, and summary statistics across various organisms, tissues, and diseases. It is ideal for large-scale, memory-efficient biological research and model training without full dataset downloads.