GrayCodeAI/starling - 3페이지

research-simpo

Simple Preference Optimization for LLM alignment. Reference-free alternative to DPO with better performance (+6.4 points on AlpacaEval 2.0). No reference model needed, more efficient than DPO. Use ...

원문 언어: 영어

research-skypilot

Multi-cloud orchestration for ML workloads with automatic cost optimization. Use when you need to run training or batch jobs across multiple clouds, leverage spot instances with auto-recovery, or o...

원문 언어: 영어

research-slime

Provides guidance for LLM post-training with RL using slime, a Megatron+SGLang framework. Use when training GLM models, implementing custom data generation workflows, or needing tight Megatron-LM i...

원문 언어: 영어

research-speculative-decoding

Accelerate LLM inference using speculative decoding, Medusa multiple heads, and lookahead decoding techniques. Use when optimizing inference speed (1.5-3.6× speedup), reducing latency for real-time...

원문 언어: 영어

research-stable-diffusion

State-of-the-art text-to-image generation with Stable Diffusion models via HuggingFace Diffusers. Use when generating images from text prompts, performing image-to-image translation, inpainting, or...

원문 언어: 영어

research-swanlab

Provides guidance for experiment tracking with SwanLab. Use when you need open-source run tracking, local or self-hosted dashboards, and lightweight media logging for ML workflows.

원문 언어: 영어

research-systems-paper-writing

Comprehensive guide for writing systems papers targeting OSDI, SOSP, ASPLOS, NSDI, and EuroSys. Provides paragraph-level structural blueprints, writing patterns, venue-specific checklists, reviewer...

원문 언어: 영어

research-tensorboard

Visualize training metrics, debug models with histograms, compare experiments, visualize model graphs, and profile performance with TensorBoard - Google's ML visualization toolkit

원문 언어: 영어

research-tensorrt-llm

Optimizes LLM inference with NVIDIA TensorRT for maximum throughput and lowest latency. Use for production deployment on NVIDIA GPUs (A100/H100), when you need 10-100x faster inference than PyTorch...

원문 언어: 영어

research-torchforge

Provides guidance for PyTorch-native agentic RL using torchforge, Meta's library separating infra from algorithms. Use when you want clean RL abstractions, easy algorithm experimentation, or scalab...

원문 언어: 영어

research-torchtitan

Provides PyTorch-native distributed LLM pretraining using torchtitan with 4D parallelism (FSDP2, TP, PP, CP). Use when pretraining Llama 3.1, DeepSeek V3, or custom models at scale from 8 to 512+ G...

원문 언어: 영어

research-transformer-lens

Provides guidance for mechanistic interpretability research using TransformerLens to inspect and manipulate transformer internals via HookPoints and activation caching. Use when reverse-engineering...

원문 언어: 영어

research-trl-fine-tuning

Fine-tune LLMs using reinforcement learning with TRL - SFT for instruction tuning, DPO for preference alignment, PPO/GRPO for reward optimization, and reward model training. Use when need RLHF, ali...

원문 언어: 영어

research-unsloth

Expert guidance for fast fine-tuning with Unsloth - 2-5x faster training, 50-80% less memory, LoRA/QLoRA optimization

원문 언어: 영어

research-verl

Provides guidance for training LLMs with reinforcement learning using verl (Volcano Engine RL). Use when implementing RLHF, GRPO, PPO, or other RL algorithms for LLM post-training at scale with fle...

원문 언어: 영어

research-vllm

Serves LLMs with high throughput using vLLM's PagedAttention and continuous batching. Use when deploying production LLM APIs, optimizing inference latency/throughput, or serving models with limited...

원문 언어: 영어

research-weights-and-biases

Track ML experiments with automatic logging, visualize training in real-time, optimize hyperparameters with sweeps, and manage model registry with W&B - collaborative MLOps platform

원문 언어: 영어

research-whisper

OpenAI's general-purpose speech recognition model. Supports 99 languages, transcription, translation to English, and language identification. Six model sizes from tiny (39M params) to large (1550M ...

원문 언어: 영어

analyzing-azure-activity-logs-for-threats

Queries Azure Monitor activity logs and sign-in logs via azure-monitor-query to detect suspicious administrative operations, impossible travel, privilege escalation, and resource modifications. Bui...

원문 언어: 영어

analyzing-cloud-storage-access-patterns

Detect abnormal access patterns in AWS S3, GCS, and Azure Blob Storage by analyzing CloudTrail Data Events, GCS audit logs, and Azure Storage Analytics. Identifies after-hours bulk downloads, acces...

원문 언어: 영어

auditing-aws-s3-bucket-permissions

Systematically audit AWS S3 bucket permissions to identify publicly accessible buckets, overly permissive ACLs, misconfigured bucket policies, and missing encryption settings using AWS CLI, S3audit...

원문 언어: 영어

auditing-azure-active-directory-configuration

Auditing Microsoft Entra ID (Azure Active Directory) configuration to identify risky authentication policies, overly permissive role assignments, stale accounts, conditional access gaps, and guest ...

원문 언어: 영어

auditing-cloud-with-cis-benchmarks

This skill details how to conduct cloud security audits using Center for Internet Security benchmarks for AWS, Azure, and GCP. It covers interpreting CIS Foundations Benchmark controls, running aut...

원문 언어: 영어

auditing-gcp-iam-permissions

Auditing Google Cloud Platform IAM permissions to identify overly permissive bindings, primitive role usage, service account key proliferation, and cross-project access risks using gcloud CLI, Poli...

원문 언어: 영어

aws-agentic-ai

AWS Bedrock AgentCore comprehensive expert for deploying and managing all AgentCore services including Gateway, Runtime, Memory, Identity, Code Interpreter, Browser, and Observability.

원문 언어: 영어

aws-cost-operations

AWS cost optimization, monitoring, and operational best practices with integrated MCP servers for billing analysis, cost estimation, observability, and security assessment.

원문 언어: 영어

aws-serverless-eda

AWS serverless and event-driven architecture expert based on Well-Architected Framework. Covers Lambda, API Gateway, DynamoDB, Step Functions, EventBridge, SQS, SNS, and patterns.

원문 언어: 영어

azure-ai-ms

Use for Azure AI: Search, Speech, OpenAI, Document Intelligence. Helps with search, vector/hybrid search, speech-to-text, text-to-speech, transcription, OCR. WHEN: AI Search, query search, vector s...

원문 언어: 영어

azure-ai

Use for Azure AI: Search, Speech, OpenAI, Document Intelligence. Helps with search, vector/hybrid search, speech-to-text, text-to-speech, transcription, OCR. WHEN: AI Search, query search, vector s...

원문 언어: 영어

azure-aigateway-ms

Configure Azure API Management as an AI Gateway for AI models, MCP tools, and agents. WHEN: semantic caching, token limit, content safety, load balancing, AI model governance, MCP rate limiting, ja...

원문 언어: 영어

azure-aigateway

Configure Azure API Management as an AI Gateway for AI models, MCP tools, and agents. WHEN: semantic caching, token limit, content safety, load balancing, AI model governance, MCP rate limiting, ja...

원문 언어: 영어

azure-cloud-migrate-ms

Assess and migrate cross-cloud workloads to Azure. Generates assessment reports and converts code from AWS, GCP, or other providers to Azure services. WHEN: migrate Lambda to Azure Functions, migra...

원문 언어: 영어

azure-cloud-migrate

Assess and migrate cross-cloud workloads to Azure. Generates assessment reports and converts code from AWS, GCP, or other providers to Azure services. WHEN: migrate Lambda to Azure Functions, migra...

원문 언어: 영어

azure-compliance-ms

Comprehensive Azure compliance and security auditing capabilities including best practices assessment, Key Vault expiration monitoring, and resource configuration validation. WHEN: compliance scan,...

원문 언어: 영어

azure-compliance

Comprehensive Azure compliance and security auditing capabilities including best practices assessment, Key Vault expiration monitoring, and resource configuration validation. WHEN: compliance scan,...

원문 언어: 영어

azure-compute-ms

Recommend Azure VM sizes, VM Scale Sets (VMSS), and configurations based on workload requirements, performance needs, and budget constraints. No Azure account required — uses public documentation a...

원문 언어: 영어

azure-compute

Recommend Azure VM sizes, VM Scale Sets (VMSS), and configurations based on workload requirements, performance needs, and budget constraints. No Azure account required — uses public documentation a...

원문 언어: 영어

azure-cost-optimization-ms

Identify and quantify cost savings across Azure subscriptions by analyzing actual costs, utilization metrics, and generating actionable optimization recommendations. USE FOR: optimize Azure costs, ...

원문 언어: 영어

azure-cost-optimization

Identify and quantify cost savings across Azure subscriptions by analyzing actual costs, utilization metrics, and generating actionable optimization recommendations. USE FOR: optimize Azure costs, ...

원문 언어: 영어

azure-deploy-ms

Execute Azure deployments for ALREADY-PREPARED applications that have existing .azure/plan.md and infrastructure files. DO NOT use this skill when the user asks to CREATE a new application — use az...

원문 언어: 영어