🎯Skills137
Assists AI researchers in drafting, structuring, and generating machine learning research papers with academic writing best practices and technical precision.
Structured ideation frameworks for discovering high-impact research directions with 10 complementary lenses (384 lines). Part of orchestra-research/ai-research-skills.
vLLM serving skill from Orchestra Research for deploying and serving large language models with high throughput and low latency.
An academic plotting skill from AI Research Skills, a library of 87 skills across 22 categories enabling AI agents to autonomously conduct AI research from idea to paper, covering model architecture, training, evaluation, and more.
Generates high-quality, customizable AI images from text prompts using advanced Stable Diffusion models with fine-tuned control over style and content.
Prompt guard skill from the AI Research Skills library for detecting and preventing prompt injection, jailbreaks, and adversarial inputs.
A Claude Code skill providing cognitive science frameworks (bisociation, structure-mapping, constraint manipulation) for generating genuinely novel research ideas.
A skill from the AI Research Engineering Skills library that teaches AI coding agents how to use Hugging Face Accelerate for distributed training and mixed-precision optimization of ML models.
A skill from the AI Research Engineering Skills Library for quantizing machine learning models using bitsandbytes, part of a comprehensive open-source collection of AI research engineering skills for AI agents.
A knowledge distillation skill from the AI Research Engineering Skills Library, the most comprehensive open-source collection of AI research engineering skills for AI agents.
A skill from the AI Research Engineering Skills Library that provides guidance on using DeepSpeed for distributed training and optimization of large-scale AI models.
A PyTorch Lightning skill from the AI Research Engineering Skills Library, providing patterns for structuring and scaling deep learning training workflows using the PyTorch Lightning framework.
A whisper skill from AI Research Skills by Orchestra Research, providing skills for academic research, paper analysis, and scientific workflows.
A verl rl training skill from AI Research Skills by Orchestra Research, providing skills for academic research, paper analysis, and scientific workflows.
A skill from the AI Research Engineering Skills library that teaches AI coding agents how to implement GGUF quantization for converting and compressing large language models into the GGUF format for efficient inference.
A audiocraft audio generation skill from AI Research Skills by Orchestra Research, providing skills for academic research, paper analysis, and scientific workflows.
A skill from the AI Research Engineering Skills Library for training large language models with Megatron, providing patterns for distributed training at scale using NVIDIA's Megatron-LM framework.
A skill from the AI Research Engineering Skills library that teaches AI coding agents how to implement HQQ (Half-Quadratic Quantization) for efficient model compression with minimal accuracy loss.
A pyvene interventions skill from AI Research Skills by Orchestra Research, providing skills for academic research, paper analysis, and scientific workflows.
A skill from the AI Research Engineering Skills Library providing guidance on AWQ (Activation-aware Weight Quantization) techniques for compressing and optimizing large language models.
A skill from the AI Research Engineering Skills library that teaches AI coding agents how to implement LlamaIndex for building data-augmented LLM applications with structured data indexing and retrieval.
A modal serverless gpu skill from AI Research Skills by Orchestra Research, providing skills for academic research, paper analysis, and scientific workflows.
A skill from the AI Research Engineering Skills library that teaches AI coding agents how to work with Sentence Transformers for generating text embeddings and semantic similarity tasks.
Hugging Face Tokenizers skill from the AI Research Engineering skills library, providing guidance for text tokenization workflows used in NLP and language model development.
A Flash Attention optimization skill from the AI Research Engineering Skills Library, providing guidance on optimizing attention mechanisms for efficient transformer inference as part of 83 AI research skills.
A sparse autoencoder training skill from the AI Research Engineering Skills Library, providing patterns for training sparse autoencoders for mechanistic interpretability and feature extraction in neural networks.
A long-context handling skill from the AI Research Engineering Skills Library, the most comprehensive open-source collection of AI research engineering skills for AI agents.
A skill from the AI Research Engineering Skills library that teaches AI coding agents how to implement Meta's Segment Anything Model (SAM) for image segmentation tasks in multimodal AI research.
A speculative decoding skill from the AI Research Engineering Skills Library, providing techniques for accelerating large language model inference using speculative decoding methods.
A skill from the AI Research Engineering Skills library that teaches AI coding agents how to implement LangSmith observability for monitoring and tracing LLM application performance.
The most comprehensive open-source library of AI research engineering skills for AI agents, covering interpretability, transformer analysis, and machine learning research workflows.
A constitutional ai skill from AI Research Skills by Orchestra Research, providing skills for academic research, paper analysis, and scientific workflows.
A SkyPilot multi-cloud orchestration skill from the AI Research Engineering Skills Library, the most comprehensive open-source library of AI research engineering skills for AI agents.
A blip 2 vision language skill from AI Research Skills by Orchestra Research, providing skills for academic research, paper analysis, and scientific workflows.
A nnsight remote interpretability skill from AI Research Skills by Orchestra Research, providing skills for academic research, paper analysis, and scientific workflows.
A skill from the AI Research Engineering Skills Library providing Phoenix observability integration for AI research workflows, part of a comprehensive open-source collection of AI research engineering skills for AI agents.
A model merging skill from the AI Research Engineering Skills Library, a comprehensive open-source collection of AI research engineering skills for AI agents.
A skill from the AI Research Engineering Skills library that teaches AI coding agents how to implement Pinecone vector database for similarity search, RAG pipelines, and embedding storage in AI applications.
Systematically assess machine learning code generation models by benchmarking performance, identifying strengths/weaknesses, and generating comparative metrics.
Model pruning skill from the AI Research Engineering skills library, providing guidance on neural network model compression and optimization techniques for AI research workflows.
The most comprehensive open-source skills library enabling AI agents to autonomously conduct AI research from idea to paper, with 87 skills across 22 categories including model architecture, fine-tuning, distributed training, inference, and safety.
Provides Ray Train distributed training patterns, part of the most comprehensive open-source AI research engineering skills library for AI agents.
A skill from the Orchestra AI Research Skills library that teaches AI agents to use SwanLab for experiment tracking, enabling logging of metrics, hyperparameters, and visualizations during machine learning research workflows.
Skill for distributed data processing with Ray Data, covering dataset pipelines, parallel transformations, and scalable data loading for ML training workflows.
MLflow experiment tracking and model management skill from Orchestra Research, part of the most comprehensive open-source AI research engineering skills library with 83 skills.
FAISS vector search skill from Orchestra Research for efficient similarity search and dense vector clustering in AI research workflows.
Part of a 98-skill AI research library enabling agents to autonomously conduct research from idea to paper. Covers the full research lifecycle including model architecture, fine-tuning, distributed training, inference, evaluation, safety, multimodal, and MLOps across frameworks like Megatron-LM, vLLM, and TRL.
TensorRT-LLM skill from Orchestra Research for optimizing and deploying large language models with NVIDIA TensorRT acceleration.
Guides AI agents through writing systems-oriented ML research papers, part of a comprehensive 87-skill library covering the full AI research lifecycle from literature survey and ideation through experiment execution to paper writing.
A pytorch fsdp2 skill from AI Research Skills by Orchestra Research, providing skills for academic research, paper analysis, and scientific workflows.
The most comprehensive open-source skills library enabling AI agents to autonomously conduct AI research from idea to paper, including conference talk presentation.
Part of the Orchestra Research AI Research Skills library, a comprehensive collection of 87 skills across 22 categories enabling AI agents to autonomously conduct AI research from idea to paper. Covers model architecture, fine-tuning, distributed training, inference, RAG, multimodal, agents, mechanistic interpretability, and more.
Part of Orchestra Research's comprehensive AI research skills library enabling agents to autonomously conduct AI research from idea to paper. Provides guidance on fine-tuning OpenVLA models using Orthogonal Fine-Tuning (OFT) for vision-language-action model adaptation.
Part of a comprehensive AI research skills library with 87 skills across 22 categories, enabling AI agents to autonomously conduct AI research from literature survey and idea generation through experiment execution to paper writing.
An AI research skill that performs semantic epistemic review of research artifacts, scoring six dimensions of research rigor (evidence relevance, falsifiability, scope, coherence, exploration integrity, methodology) with severity-ranked findings. Part of the Orchestra Research AI Research Skills Library.
A post-task research recorder that runs at session end to extract decisions, experiments, dead ends, and pivots from conversation history into an Agent-Native Research Artifact (ARA) directory with user-vs-AI provenance tags.
A Claude Code skill that compiles any research input (PDF papers, GitHub repos, experiment logs, raw notes) into a structured Agent-Native Research Artifact with claims, exploration graph, grounded evidence, and code stubs, part of the Orchestra Research AI research skills library.
A skill from the AI Research Engineering Skills library that teaches AI coding agents how to implement LangChain for building LLM-powered applications with chains, agents, and retrieval-augmented generation.
A unsloth skill from AI Research Skills by Orchestra Research, providing skills for academic research, paper analysis, and scientific workflows.
A CrewAI multi-agent skill from the AI Research Engineering Skills Library, providing patterns for building multi-agent systems using the CrewAI framework for collaborative AI research workflows.
A skill from the AI Research Engineering Skills library that teaches AI coding agents how to implement Chroma vector database for embedding storage, similarity search, and retrieval-augmented generation pipelines.
Part of the AI Research Skills Library by Orchestra Research, a comprehensive collection of 87 skills enabling AI agents to autonomously conduct AI research from idea to paper. This skill focuses on ML training recipes within the broader library covering model architecture, fine-tuning, distributed training, inference, and more.
A LlamaGuard skill from the AI Research Engineering Skills Library, providing guidance on using Meta's LlamaGuard models for AI safety classification and content moderation in research workflows.
An SGLang skill from the AI Research Engineering Skills Library, providing research engineering knowledge for AI agents working with the SGLang framework.
A evaluating llms harness skill from AI Research Skills by Orchestra Research, providing skills for academic research, paper analysis, and scientific workflows.
A Qdrant vector search skill from the AI Research Engineering Skills Library, part of a comprehensive open-source collection of AI research engineering skills for AI agents.
An Instructor skill from the AI Research Engineering Skills Library, providing guidance on structured output extraction from language models as part of 83 research skills across 20 categories.
A weights and biases skill from AI Research Skills by Orchestra Research, providing skills for academic research, paper analysis, and scientific workflows.
A skill from the AI Research Engineering Skills library that teaches AI coding agents how to implement NVIDIA NeMo Guardrails for LLM safety, covering jailbreak detection, input/output validation, and hallucination detection.
A tensorboard skill from AI Research Skills by Orchestra Research, providing skills for academic research, paper analysis, and scientific workflows.
An Outlines skill from the AI Research Engineering Skills Library, providing guidance on structured text generation and constrained decoding for language models as part of 83 AI research skills across 20 categories.
A DSPy skill from the AI Research Engineering Skills Library, providing patterns for building and optimizing language model programs using the DSPy framework as part of a comprehensive AI research skills collection.
NeMo Curator skill from the AI Research Engineering skills library, providing guidance for data curation workflows used in training large language models.
A llama cpp skill from AI Research Skills by Orchestra Research, providing skills for academic research, paper analysis, and scientific workflows.
A skill from the AI Research Engineering Skills library that teaches AI coding agents how to implement GRPO (Group Relative Policy Optimization) for reinforcement learning training of language models.
A clip skill from AI Research Skills by Orchestra Research, providing skills for academic research, paper analysis, and scientific workflows.
An OpenRLHF training skill from the AI Research Engineering Skills Library, providing guidance on reinforcement learning from human feedback training techniques as part of 83 AI research skills across 20 categories.
A axolotl skill from AI Research Skills by Orchestra Research, providing skills for academic research, paper analysis, and scientific workflows.
A sentencepiece skill from AI Research Skills by Orchestra Research, providing skills for academic research, paper analysis, and scientific workflows.
A llama factory skill from AI Research Skills by Orchestra Research, providing skills for academic research, paper analysis, and scientific workflows.
A torchforge rl training skill from AI Research Skills by Orchestra Research, providing skills for academic research, paper analysis, and scientific workflows.
A skill from the AI Research Engineering Skills library that teaches AI coding agents how to implement MILES (Model-based Imitation Learning from Extreme States) reinforcement learning training for robust policy optimization.
A Guidance skill from the AI Research Engineering Skills Library, providing guidance on constrained generation and structured output from language models as part of 83 research skills across 20 categories.
A skill from the AI Research Engineering Skills library that teaches AI coding agents how to implement GPTQ (GPU-optimized post-training quantization) for compressing large language models while maintaining accuracy.
A skill from the AI Research Engineering Skills library that teaches AI coding agents how to implement Parameter-Efficient Fine-Tuning (PEFT) techniques like LoRA and QLoRA for adapting large language models.
A skill from the AI Research Engineering Skills library that teaches AI coding agents how to implement Mixture of Experts (MoE) training for building sparse, efficient large language models.
Skill for Lambda Labs GPU cloud operations from the AI Research Engineering skills library, covering AI research infrastructure including GPU cloud management for machine learning workloads.
A SimPO training skill from the AI Research Engineering Skills Library, providing patterns for implementing Simple Preference Optimization training for language model alignment.
A SLIME reinforcement learning training skill from the AI Research Engineering Skills Library, providing guidance on RL training techniques as part of 83 AI research skills across 20 categories.
A distributed llm pretraining torchtitan skill from AI Research Skills by Orchestra Research, providing skills for academic research, paper analysis, and scientific workflows.
A fine tuning with trl skill from AI Research Skills by Orchestra Research, providing skills for academic research, paper analysis, and scientific workflows.
AutoGPT agents skill from the AI Research Engineering skills library, providing specialized knowledge for working with autonomous AI agent frameworks in research contexts.
A llava skill from AI Research Skills by Orchestra Research, providing skills for academic research, paper analysis, and scientific workflows.
NanoGPT skill from the AI Research Engineering skills library, providing guidance for training small-scale GPT language models for AI research and experimentation.
An RWKV architecture skill from the AI Research Engineering Skills Library, a comprehensive open-source collection of AI research engineering skills for AI agents.
A mamba architecture skill from AI Research Skills by Orchestra Research, providing skills for academic research, paper analysis, and scientific workflows.
A nemo evaluator sdk skill from AI Research Skills by Orchestra Research, providing skills for academic research, paper analysis, and scientific workflows.
A implementing llms litgpt skill from AI Research Skills by Orchestra Research, providing skills for academic research, paper analysis, and scientific workflows.
AI research engineering skill providing guidance on MILES reinforcement learning training, from a comprehensive open-source library of research skills for AI agents.
A reinforcement learning training skill for the SLIME framework (THUDM's Megatron+SGLang framework powering GLM-4.x models), part of the AI Research Skills library covering 83 skills across post-training, distributed training, and optimization.
Part of an 82-skill AI research engineering library covering 20 categories including model architecture, fine-tuning, distributed training, inference, RAG, multimodal, and MLOps, enabling coding agents to conduct AI research experiments and training pipelines.
AI research engineering skill providing guidance on distributed LLM pretraining using TorchTitan, from a comprehensive open-source library of research skills for AI agents.
AI research engineering skill for using Unsloth to perform efficient LLM fine-tuning, part of a comprehensive library of 82 skills that enable coding agents to conduct AI research experiments.
Provides AI research engineering guidance for building applications with LangChain, part of the Orchestra Research AI Research Skills library covering agent frameworks and LLM application development.
Provides AI research engineering guidance for DSPy, a framework for programming language model pipelines with automatic prompt optimization, part of the Orchestra Research AI Research Skills library covering prompt engineering techniques.
Provides AI research engineering guidance for Group Relative Policy Optimization (GRPO) reinforcement learning training, part of the Orchestra Research AI Research Skills library covering post-training techniques for language models.
AI research skill for implementing NVIDIA NeMo Guardrails, enabling programmable guardrails with the Colang language to control LLM behavior and enforce safety policies.
AI research skill that provides expert-level guidance for working with OpenAI's CLIP vision-language model, covering zero-shot image classification, text-image similarity, and multimodal embeddings.
AI research skill for implementing Meta's LlamaGuard, a safety classifier that evaluates LLM inputs and outputs for harmful content, part of the AI Research Skills library.
AI research skill for using SGLang, a high-performance inference serving framework with RadixAttention that provides 5-10x faster structured generation for AI agents and complex workflows.
AI research skill for using Microsoft Research's Guidance library, which enables constrained generation with regex and grammars to produce structured and reliable LLM outputs.
Provides AI research engineering guidance for GPTQ post-training quantization of large language models, part of the Orchestra Research AI Research Skills library covering optimization techniques for reducing model size and inference costs.
AI research skill providing expert guidance on OpenRLHF, a full RLHF (Reinforcement Learning from Human Feedback) training pipeline built on Ray and vLLM for scalable post-training of language models.
Provides AI research engineering guidance for llama.cpp, a C++ inference engine for running large language models locally, part of the Orchestra Research AI Research Skills library covering inference optimization techniques.
Provides AI research engineering guidance for using SentencePiece tokenization, part of the Orchestra Research AI Research Skills library covering tokenization techniques for language model training and inference.
Provides AI research engineering guidance for MLflow experiment tracking and model management, part of the Orchestra Research AI Research Skills library covering MLOps workflows for AI research.
AI research skill for using LLaMA-Factory, a WebUI-based no-code fine-tuning framework that simplifies training and customizing large language models without writing code.
AI research skill for working with NanoGPT, Andrej Karpathy's educational GPT implementation in approximately 300 lines of code, designed for learning and experimenting with transformer architectures.
Part of the AI Research Skills library by Orchestra Research, providing guidance for Mixture of Experts (MoE) model training including architecture setup, expert routing, and load balancing.
Provides expert-level guidance on Chroma, an open-source vector database for building RAG (Retrieval-Augmented Generation) applications. Part of the AI Research Engineering Skills library covering 82+ skills for AI research, including embedding storage, similarity search, and retrieval pipelines.
Event-driven investment strategy framework via Longbridge — identifies and analyzes corporate events (M&A, spinoffs, buybacks, index rebalancing, lockup expiry) that create pricing dislocations, with event identification, sentiment scoring, and position sizing.
Part of the AI Research Skills library by Orchestra Research, providing guidance for fine-tuning language models using TRL (Transformer Reinforcement Learning) with SFT, RLHF, and DPO methods.
Provides expert-level guidance on NVIDIA NeMo Curator for AI dataset preparation and curation. Part of the AI Research Engineering Skills library covering 83+ skills across the full AI research lifecycle, from data processing to model deployment.
Provides expert-level guidance on implementing LLMs using LitGPT (Lightning AI), covering 20+ clean LLM implementations with production training recipes. Part of the AI Research Engineering Skills library spanning model architecture, fine-tuning, distributed training, and more.
AI research skill that provides expert guidance on Mamba state-space models, which achieve O(n) sequence complexity and up to 5x faster inference than Transformers, ideal for long-context tasks.
Provides AI research engineering guidance for TensorRT-LLM, NVIDIA's high-performance inference library for large language models, part of the Orchestra Research AI Research Skills library covering inference optimization on GPU hardware.
AI research skill providing expert guidance on Outlines, a structured text generation library that uses finite state machines (FSM) for zero-overhead constrained output from language models.
Part of the AI Research Skills library by Orchestra Research, providing guidance for SimPO (Simple Preference Optimization) training workflows for aligning language models without a reference model.
AI research skill providing expert guidance on Weights & Biases (W&B) for experiment tracking, hyperparameter sweeps, artifact management, and model registry in machine learning workflows.
An AI research skill for TensorBoard visualization, profiling, embeddings, and scalars/images monitoring, part of a comprehensive 83-skill library for AI research engineering spanning model training to deployment.
An AI research skill for using the Instructor library to produce structured LLM outputs with Pydantic validation, part of a comprehensive 83-skill library covering the full AI research engineering lifecycle.
AI research engineering skill providing guidance on evaluating LLMs using evaluation harnesses, from a comprehensive open-source library of research skills for AI agents.
Part of the AI Research Skills library by Orchestra Research, providing guidance for using Axolotl, a tool for streamlined fine-tuning of language models with multi-GPU support and various training methods.
AI research engineering skill providing guidance on LLaVA (Large Language-and-Vision Assistant) implementation, from a comprehensive open-source library of research skills for AI agents.
Part of the AI Research Skills library by Orchestra Research, providing guidance for building and using FAISS (Facebook AI Similarity Search) vector indexes for efficient similarity search in AI applications.
A comprehensive open-source skills library that enables AI agents to autonomously conduct AI research, from generating ideas to writing papers. Provides a full research workflow pipeline for AI-powered academic work.
Skill
🔌Plugins20
Plugin
Plugin
Plugin
Plugin
Plugin
A tokenization skill from the AI Research Engineering Skills Library, which offers 83 skills across 20 categories covering model architecture, fine-tuning, inference, and other AI research areas.
Plugin
Plugin
Plugin
AI research skill for writing publication-ready ML papers for top conferences (NeurIPS, ICML, ICLR, ACL, AAAI, COLM) with LaTeX templates and citation verification.
Plugin
Plugin
Plugin
Plugin
A collection of 4 AI safety and alignment research skills covering Constitutional AI, LlamaGuard safety classifier, NeMo programmable guardrails with Colang, and Meta's Prompt Guard injection detector — part of Orchestra Research's AI research engineering skills.
Plugin
A collection of 7 multimodal AI research skills covering CLIP, Whisper, LLaVA, BLIP-2, SAM, Stable Diffusion, and AudioCraft — part of Orchestra Research's 83 AI research engineering skills for coding agents.
Plugin
Prompt-engineering category of the AI Research Engineering Skills library — 4 skills covering DSPy (declarative prompt programming), Instructor (Pydantic-validated structured outputs), Guidance (regex/grammar-constrained generation), and Outlines (FSM-based structured text).
Emerging-techniques category of the AI Research Engineering Skills library — 6 skills covering Mixture-of-Experts training, Model Merging (TIES/DARE/SLERP via mergekit), Long Context (RoPE/YaRN/ALiBi), Speculative Decoding, Knowledge Distillation, and Model Pruning.