Skill UI

Comprehensive guide to LLM Operations (LLMOps). Covers advanced topics like Retrieval-Augmented Generation (RAG), vector database management (Pinecone, Chroma), embeddings, fine-tuning, prompt engineering, and cost optimization. Essential for bridging the gap between AI prototypes and reliable, scalable, production-grade AI systems.

This skill automates the Design Space Exploration (DSE) process, crucial for computer architecture and Electronic Design Automation (EDA). It runs complex simulations, analyzes performance metrics, and systematically tunes parameters (e.g., cache size, pipeline width) across a defined parameter space. It iterates until the objective metric is optimized or a defined timeout is reached, making it ideal for complex hardware tuning and system design.

This skill provides comprehensive guidance on deploying and tuning Web Application Firewall (WAF) rules across major cloud platforms, including AWS, Azure, and Cloudflare. It details strategies to protect cloud-hosted applications against OWASP Top 10 attacks, implement rate limiting for brute force protection, configure geo-blocking, and refine rules to minimize false positives.

Guides fine-tuning and evaluation of OpenVLA-OFT/OFT+ for robot action generation, covering LoRA adaptation, continuous action heads, FiLM conditioning, and LIBERO/ALOHA benchmarks plus deployment considerations.

Provides end-to-end guidance for fine-tuning and serving Physical Intelligence’s OpenPI pi0 family (pi0, pi0-fast, pi0.5) with both JAX and PyTorch backends, covering dataset prep, norm stats, checkpoint conversion, and low-latency WebSocket policy inference for ALOHA, DROID, and LIBERO tasks.

This skill provides a managed cloud environment for training and fine-tuning advanced vision models. Users can build and customize models for object detection (e.g., DETR, YOLOS), image classification, and segmentation (SAM/SAM2). By leveraging Hugging Face Jobs, training is executed on powerful cloud GPUs without requiring local GPU setup, and all resulting models are automatically saved to the Hugging Face Hub.

Utilize Modal, a serverless GPU cloud, for running complex machine learning workloads, including training, fine-tuning, and inference. This tool provides a zero-config environment, eliminating the need for SSH setup or Docker management. It features automatic scale-to-zero billing and local-first execution, making high-powered GPU compute accessible for debugging and small-scale experiments.

This skill provides a comprehensive framework for optimizing, improving, and debugging Large Language Model (LLM) prompts. It leverages production trace data, system evaluations, and human annotations to extract prompts from spans. Users can run a data-driven optimization loop, ensuring the AI responses meet desired quality standards. Ideal for prompt tuning, system prompt refinement, and improving overall AI output quality.

This comprehensive guide provides advanced techniques to diagnose and significantly boost ClickHouse query performance. It covers analyzing slow queries using system tables, optimizing ORDER BY keys, implementing data skipping indexes (Bloom, Set, MinMax), utilizing projections for pre-aggregation, and fine-tuning server settings and materialized views. Essential for high-volume OLAP environments.

This guide details how to set up and execute large-scale, distributed GPU training jobs on the CoreWeave platform. It covers both single-node multi-GPU setups and multi-node training utilizing PyTorch DDP and Kubernetes resources (A100/H100). This workflow is ideal for fine-tuning large language models (LLMs) or running complex deep learning models that require high-performance, shared-storage compute clusters.

This comprehensive skill manages the entire computer vision pipeline, from custom dataset annotation to model deployment. Users can fine-tune YOLO models on proprietary COCO-formatted datasets using an agent-driven workflow. Key features include hardware-aware training (supporting CUDA, MPS, ROCm), automatic export to optimized formats (TensorRT, CoreML, OpenVINO, ONNX), and seamless, one-click deployment as an active detection skill.

A comprehensive reference guide for common errors encountered when interacting with the Together AI API. This includes error codes and solutions for inference, fine-tuning, and model deployment, covering authentication, rate limits, validation failures, and billing issues. Essential for developers integrating OpenAI-compatible LLM services.