데이터·AI 스킬

Build and evaluate Mamba models for long sequences using state-space models, which summarize past inputs in an internal state. Covers setup, pretrained text generation, differences between Mamba-1 and Mamba-2, and speed comparisons with Transformers.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Generate text with pretrained Mamba models
• Compare Mamba-1 and Mamba-2 configurations
• Benchmark generation speed against Transformers

Design and implement machine learning systems for production use. Cover data preparation, training, prediction services, model comparisons, monitoring, and retraining, with attention to reliability, response time, resource costs, and both technical and business evaluation metrics.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Designing a real-time recommendation service
• Comparing model versions with A/B testing
• Detecting changes in data and model performance

Use SAM to identify object regions in images from points, boxes, or existing masks. The skill provides implementation examples for transparent cutouts, automatic mask generation, annotation tools, and model selection based on available resources.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Extracting objects with transparent backgrounds
• Building annotation tools with point or box selection
• Generating object masks automatically

Deploy language models with vLLM through an OpenAI-compatible API or offline batch processing. Tune response speed, throughput, and GPU memory use with quantization, which reduces numerical precision, and execution across multiple GPUs.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Deploy an OpenAI-compatible language model API
• Generate text from many prompts in a batch
• Serve quantized models with limited GPU memory

Use TensorBoard to visualize training loss and accuracy, compare experiments, and inspect model structure. Includes PyTorch and TensorFlow logging examples for examining weight distributions, predictions, and performance bottlenecks during machine learning training.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Comparing loss and accuracy across training runs
• Inspecting weight and gradient distributions
• Exploring embeddings in two or three dimensions

Use LitGPT to explore language model implementations and train models on custom data. Covers pretrained model loading, LoRA fine-tuning that updates a small set of parameters, pretraining, and preparing trained models for deployment.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Studying language model architecture through readable code
• Fine-tuning a model on custom instruction data
• Configuring LoRA training for a single GPU

Investigate how transformer language models produce predictions with TransformerLens. This skill guides inspection of intermediate values and attention patterns, plus experiments that replace internal values to test their effects and identify components involved in model behavior.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Inspecting attention patterns and intermediate values
• Testing output changes by replacing internal activations
• Analyzing attention head contributions to predictions

Set up LLM serving with structured responses, including JSON and outputs restricted to specified patterns. Covers reusing shared prompt calculations, tool calling, and multi-turn conversations through SGLang, with examples for server setup and application integration.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Extract information into JSON with a defined schema
• Build agents that call tools
• Reuse shared prompts across chat requests

Integrate Flash Attention into transformer models to reduce processing time and GPU memory use for long sequences. Covers PyTorch integration, advanced attention options, and benchmarks that compare speed, memory use, and output accuracy.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Speeding up transformer training on long sequences
• Reducing GPU memory use during attention
• Adding Flash Attention to a PyTorch model

Set up Langfuse to record LLM application activity, costs, and response times. Manage prompt versions, organize evaluation datasets, and compare results to investigate failures and guide improvements using recorded evidence.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Investigating failures in LLM applications
• Comparing prompt versions and evaluation scores
• Monitoring LLM costs and response times

Integrate the Gemini API on Agent Platform using the Google Gen AI SDK. Guidance covers authentication, text and multimedia tasks, structured JSON responses, live voice interactions, context caching, and batch prediction across five programming languages.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Adding document and multimedia understanding to an application
• Generating JSON that matches a defined schema
• Building live voice and video interactions

Configure LLM inference and serving on NVIDIA GPUs with TensorRT-LLM. Covers quantization, which reduces numerical precision, batching requests during generation, and distributing models across GPUs to improve processing capacity and response times.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Setting up an LLM server on NVIDIA GPUs
• Configuring FP8 or INT4 quantization
• Distributing large models across multiple GPUs

Get guidance on adapting large language models with Axolotl. Covers YAML configuration files, LoRA and QLoRA methods for training fewer parameters, distributed training settings, dataset formats, and debugging training code.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Write YAML settings for model fine-tuning
• Explore training with LoRA or QLoRA
• Configure training across multiple GPUs

Run language models locally with llama.cpp on CPUs, Apple Silicon, or AMD and Intel GPUs. The skill guides model setup, memory-saving GGUF format choices, interactive chat, and serving models through an OpenAI-compatible API.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Running a language model on a CPU-only computer
• Setting up local chat on an Apple Silicon Mac
• Choosing model quantization to fit available memory

Build image captioning, visual question answering, and text-based image search with BLIP-2. The skill explains model selection, batch processing, and memory-saving settings, with Python examples for connecting image understanding to language models.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Generating captions for multiple images
• Building a question-answering system for images
• Searching an image collection with text

Design LLM applications using patterns for answers grounded in documents, agents that use tools, and prompt testing. Implementation examples also cover monitoring response quality, latency, costs, and failures during operation.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Building an assistant that answers from internal documents
• Choosing an architecture for an agent that uses tools
• Comparing and versioning prompts

Get guidance on distributed training with PyTorch FSDP, which splits model parameters across devices. The skill covers implementation, mixed precision, moving data to CPU memory, communication setup, and debugging distributed training code.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Implement distributed training with FSDP
• Review mixed precision and CPU offloading settings
• Choose a distributed communication backend

Manage the limited information an LLM can process at once. Use importance-based summaries, selective trimming, token counting, and content placement to plan what stays in context and address information loss during long conversations.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Planning how to handle conversations near the token limit
• Summarizing a long dialogue while preserving key information
• Reviewing the placement of important instructions and facts

Extend the amount of text a language model can process at once. Compares RoPE, YaRN, ALiBi, and position interpolation, with implementation examples, additional training guidance, and memory optimization approaches for long document processing.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Extend an existing model to handle longer inputs
• Compare methods for representing token positions
• Plan additional training with long documents

Reduce LLM weight usage through pruning, which removes selected model parameters. Explore Wanda and SparseGPT for compression without retraining, choose hardware-friendly patterns for faster inference, and compare the compressed model's accuracy with the original.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Compressing an LLM without retraining
• Comparing pruning methods
• Exploring faster inference on supported hardware

Adapt machine learning training to run across multiple GPUs or machines with Ray Train. Covers changes to existing code, comparing training settings with Ray Tune, and saving checkpoints so training can resume after failures.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Adapting existing PyTorch code for multiple GPUs
• Training HuggingFace models across multiple machines
• Comparing learning rates and batch sizes with Ray Tune

Train smaller language models using knowledge distillation, where a larger model supplies examples or prediction targets. This skill covers training losses, generated training data, parameter choices, and output comparisons to support model compression and specialized student models.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Training a smaller model from a larger teacher
• Fine-tuning with teacher-generated responses
• Distilling domain knowledge into a specialized model

Helps speed up LLM text generation by comparing methods that predict and verify multiple tokens at once. Covers draft model selection, Medusa prediction heads, Lookahead Decoding settings, and deployment examples for inference services.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Reducing chatbot response latency
• Comparing methods for faster code generation
• Selecting a small draft model for a larger LLM

Build collaborating AI teams with CrewAI by defining roles, tasks, and shared context. Configure ordered execution or manager-led delegation, add memory and tools, and create workflows that branch based on events or results.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Build a research-to-writing agent workflow
• Coordinate specialists through a manager agent
• Create workflows with conditional branches