Skill UI

Guides CMOs through brand positioning, growth model selection, budget allocation, and org design for scaling companies, using diagnostic questions and projections to align marketing with revenue goals.

This skill is designed for performing local, on-premise evaluations and benchmarking of models hosted on the Hugging Face Hub. It facilitates running specialized evaluation frameworks like `inspect-ai` and `lighteval` against local hardware. Users can select optimal inference backends, including `vllm` for high throughput, or fall back to Hugging Face Transformers or `accelerate`, ensuring comprehensive smoke testing and task selection for LLMs. It focuses purely on the local execution process, separate from remote jobs or publishing workflows.

Collection of battle-tested PyTorch training recipes covering architecture selection, optimizers, LR scheduling, precision tips, and debugging to train or fine-tune LLMs, vision, diffusion, and biomedical models with high GPU throughput.

A comprehensive guide for configuring and managing Claude API access across disparate environments (Development, Staging, Production). This skill details how to implement isolated API keys, model selection strategies, and crucial spend controls using Anthropic Workspaces, ensuring secure and reliable deployment pipelines for AI applications.

This guide details advanced strategies for optimizing Claude API usage to reduce latency and cost. Techniques covered include prompt caching (significant cost reduction), strategic model selection (Haiku, Sonnet, Opus) based on task complexity, implementing streaming for perceived speed, and managing token limits. Use this guide when building high-performance, cost-efficient LLM applications that interact with large language models.

This guide outlines advanced strategies to optimize latency and throughput when integrating Anthropic's Claude API. Learn critical techniques such as streaming responses, prompt caching, model selection (Haiku, Sonnet, Opus), and request parallelization, ensuring your applications provide a low-latency, high-performance user experience.

This skill provides a comprehensive guide to optimizing costs when using AssemblyAI for speech-to-text transcription. It details model selection (Best vs. Nano), guides feature budgeting (e.g., Diarization, Sentiment Analysis), and includes code examples for cost estimation and usage tracking. Use this when managing billing, reducing operational expenses, or implementing budget alerts for high-volume audio processing.

Provides strategies to optimize costs when utilizing the Cohere API. This guide covers model tier selection (economy, standard, premium), implementing token budgets, reducing embedding storage costs (e.g., using int8), and establishing comprehensive usage trackers. It is essential for maintaining cost predictability in production AI deployments.

A comprehensive guide for configuring and integrating the Cohere API v2 across development, staging, and production environments. This pattern ensures environment-specific settings—including API keys, model selections, token limits, and caching strategies—are correctly applied. It covers best practices for secure secret management using tools like Vault, AWS Secrets Manager, and GCP Secret Manager, ensuring robust and scalable deployments.

A comprehensive guide for optimizing Cohere API v2 performance. Learn advanced strategies to improve throughput, reduce latency, and manage costs when using Chat, Embed, and Rerank endpoints. Techniques covered include dynamic model selection based on latency budgets, implementing streaming for faster perceived response times, efficient batch processing for embeddings, vector compression, and robust caching mechanisms.

This guide provides expert strategies for optimizing GPU inference performance on CoreWeave infrastructure. It covers GPU selection based on workload (LLM, image generation, training), advanced techniques like continuous batching (using vLLM), autoscaling setup (HPA), and benchmarking data. Use this to maximize GPU utilization, minimize latency, and optimize large-scale AI model deployment.

This skill provides methods to optimize ElevenLabs usage costs for Text-to-Speech (TTS) workloads. It covers model selection (e.g., using Flash/Turbo for savings), implementing character-efficient text pre-processing to reduce character count, and setting up real-time quota monitoring with budget alerts. Ideal for developers managing high-volume, cost-sensitive voice applications.