NVIDIA
- 1.2k skills
- 0 followers
- 23k repo stars
- 561 verified
- 11 hours ago last updated
- ▌ Warp Compile Time Optimizer · nvidia bundleUse when compile time or startup time is the problem in code that uses Warp: a request to improve, optimize, or cut compile times; an app that is slow to start or stalls at the first wp.launch; seconds of compiling before real work begins; JIT modules recompiling on every run or every CI job. Only applies when the code being optimized uses Warp kernels. Not for steady-state kernel runtime, memory, correctness, building Warp itself from source, or nvcc/C++ build times.
- ▌ Nemo Relay Migrate From Flow · nvidia bundleUse this skill when migrating applications, examples, integrations, documentation, manifests, or repository code from NeMo Flow to NeMo Relay across Python, Rust, Node.js, Go, C FFI, CLI, configuration, and observability surfaces.
- ▌ Paidf Curation And Retrieval · nvidia bundleUse when operating PAIDF Curation and Retrieval or NVIDIA Cosmos Curator pipelines (split, filter, caption, embed, dedup, shard, image annotate) or PAIDF Data Mining nearest-neighbor matching on Curator embeddings. Activate for Make or CLI pipeline config, GPU run preflight, FFmpeg sidecar, SAM3 keys, or Curator-to-TAO handoff. Do not use for generic ETL, vector-database RAG, model training, orchestration, or embeddings outside Cosmos Curator and PAIDF Data Mining.
- ▌ Rtvi Cv Scaffold Vss Service · nvidia bundleScaffold a standalone RTVI CV microservice that plugs into VSS Search and Alerts profiles via Kafka mdx-raw. The shipped scaffold script is a YOLO26 reference implementation (ONNX, labels, custom parser required). Use when building a new perception microservice repo, validating the VSS integration contract, extending that scaffold for segmentation frame-mask payloads, or scaffolding with placeholders before customer YOLO26 assets exist. For swapping the detector in the stock vss-rt-cv container, use rtvi-cv-customize-model instead. Live DeepStream integration cannot run until the customer-supplied ONNX, labels file, and parser library exist.
- ▌ I4h Catheter Navigation Smoke · nvidia bundleRun CPU-only fluorosim smoke tests (imports, preprocessing, CLI parsers). Use when asked to smoke-test catheter navigation in CI or without a GPU.
- ▌ Paidf Orchestration Write Dag · nvidia bundleUse when a user describes a custom PAIDF Orchestration pipeline — a specific ordered combination of stages such as augmentation only, auto-labeling only, detection+captioning only, or image attribute augmentation without full auto-labeling — that no existing DAG in airflow/dags/workflows/ covers, and asks for a new Kubernetes DAG. Also use to check that a generated or existing DAG's model/container/prompt choices match an external spec document (e.g. a PAIDF `launchable.md`).
- ▌ Nemo Relay Plugin Observability · nvidia bundleUse this skill when choosing or configuring NeMo Relay 0.6 or 0.7 observability through the built-in plugin, subscribers, or exporters, including raw ATOF events, ATIF trajectories, OpenTelemetry, OpenInference, or custom event handling.
- ▌ Doca Gpi · nvidia bundleUse this skill for hands-on DOCA GPI programming — wiring a GPU-Packet-Initiator context so a CUDA kernel drives RDMA queues directly from GPU memory without host CPU mediation. Covers picking GPI vs doca-gpunetio, the doca_gpi / domain / channel object model, the GPU-side handle handoff (doca_gpu_gpi_channel*), attaching GPU memory to a GPI domain, the domain and channel attribute objects, and debugging DOCA_ERROR_* from doca_gpi_* calls. Trigger even when the user does not explicitly mention "DOCA GPI" — implicit phrasings include "my CUDA kernel needs to post RDMA directly from GPU memory", "DOCA_ERROR_* from doca_gpi_gpu_channel_get", "how do I hand a GPU handle to my CUDA kernel", "how many channels can a GPI domain hold", or "GPU kernel driving RDMA without the host CPU on the path". Refuse and route elsewhere for the doca-gpunetio Send/Receive surface, the doca-rdma queue lifecycle, DPA-side initiation (doca-rdmi), or the CUDA programming model — those belong to other skills.
- ▌ Doca Rdmi · nvidia bundleUse this skill when the user is doing hands-on DOCA RDMI (RDMA Initiator) programming — picking doca-rdmi vs doca-rdma for an accelerator-initiated one-sided RDMA flow, standing up a doca_rdmi_connection or doca_rdmi_poster, attaching a doca_dpa_completion or doca_verbs_cq before doca_ctx_start(), retrieving the DPA-side handle for a DPA kernel, auditing whether a doca_rdmi_* symbol is EXPERIMENTAL on this DOCA, or debugging DOCA_ERROR_* returns from RDMI calls. Trigger even when the user does not say "DOCA RDMI" or "initiator" — implicit phrasings include "my DPA kernel needs to post RDMA writes to a remote responder", "DPA kernel sees no completions", "function not found on doca_rdmi_* at link time", "DOCA_ERROR_BAD_STATE from completion attach", or "the DPA posted but the work request never arrived". Refuse and route elsewhere for two-sided or host-CPU RDMA, the DPA programming model, GPU-side RDMA initiation, or general RDMA/IB/RoCE concepts — those belong to other skills.
- ▌ Warp Eval · nvidia bundleEvaluate whether an existing hot path is a credible NVIDIA Warp candidate. Use for irregular or spatial queries, particle or geometry simulation, branch-heavy loops, many small launches, host fallbacks, or large intermediates. CPU-only code and absent GPU dependencies are normal unless NVIDIA is prohibited. Exclude required cross-vendor or CPU-only deployment, vendor-lowered dense or NN layers, general Warp API questions, and already-selected Warp kernels. Contribution policy alone is not exclusion.
- ▌ Hsb Ip Def · nvidia bundleGenerate, validate, compare, or explain HSB HOLOLINK_def.svh macros. Do not use for FPGA_top.sv wrappers or packetizer-only derivation. Generation runs bundled Python scripts locally through shell commands and writes validated .svh output files after user-confirmed paths.
- ▌ Earth2studio Create Prognostic · nvidia bundleCreate Earth2Studio prognostic model wrappers that time-step weather forecasts forward, with triple-inheritance classes, tests, and documentation.
- ▌ Cuopt Multi Objective Exploration · nvidia bundleTrace and interpret the Pareto frontier across competing objectives using repeated single-objective cuOpt solves (weighted-sum and ε-constraint).
- ▌ Deepstream Sop · nvidia bundleBuild, deploy, evaluate, debug, and measure latency for a GPU-accelerated FastAPI service that detects whether operators perform assembly-line steps in order via event boundary detection and VLM classification.
- ▌ Amc Run Video Calibration · nvidia bundleCalibrate a new dataset from pre-recorded video files via the AutoMagicCalib REST API.
- ▌ Amc Run Sample Calibration · nvidia bundleRun end-to-end calibration on the bundled sample dataset against a running AMC microservice to verify the stack works before processing real data.
- ▌ Amc Setup Calibration Stack · nvidia bundleDeploy the AutoMagicCalib microservice and web UI from pre-built NGC release images using Docker Compose.
- ▌ Deepstream Profile Pipeline · nvidia bundleProfile a DeepStream pipeline with Nsight Systems and derive its configs from the measurement.
- ▌ Deepstream Generate Pipeline · nvidia bundleBuilds and validates DeepStream GStreamer pipelines through an interactive questionnaire and a BM25 retrieval engine over 270+ verified pipelines.
- ▌ Earth2studio Create Datasource · nvidia bundleCreate and validate Earth2Studio data source wrappers (DataSource, ForecastSource, DataFrameSource, ForecastFrameSource) from remote stores like S3, GCS, Azure, HTTP, or HuggingFace.
- ▌ Earth2studio Create Diagnostic · nvidia bundleCreate Earth2Studio diagnostic model wrappers for single-step data transformations, including simple derived diagnostics, packaged AutoModel diagnostics, and generative or diffusion diagnostics.
- ▌ Nemo Mbridge Perf Sequence Packing · nvidia bundleValidate and configure packed sequences and long-context training in Megatron-Bridge, distinguishing offline packed SFT for LLMs from in-batch packing for VLMs with correct context parallelism constraints.
- ▌ Vss Generate Video Report · nvidia bundleGenerates video analysis reports by routing to a VLM backend for per-clip analysis or an analytics backend for incident-range reports, with deployment profile verification and URL rewriting.
- ▌ Tao Convert Dataset Format · nvidia bundleConverts NVIDIA TAO DAFT datasets between supported formats using the `tao-daft convert` CLI.
- ▌ Tao Finetune Cosmos Reason · nvidia bundleFine-tune Cosmos Reason video QA models using supervised fine-tuning with FSDP parallelism, including dataset preparation, spec construction, and AutoML support.
- ▌ Tao Port Huggingface Model · nvidia bundleIntegrate a HuggingFace computer vision model into the NVIDIA TAO Toolkit ecosystem, covering the full pipeline from prerequisites to container testing.
- ▌ Tao Train Visual Changenet · nvidia bundleTrains, evaluates, exports, and runs inference for Visual ChangeNet models used in AOI defect detection, comparing image pairs for PASS/NO_PASS classification or change-segmentation masks.
- ▌ Vss Deploy Video Embedding · nvidia bundleDeploy and operate the VSS 3.2 GA RT-Embed Video Embedding microservice using Docker Compose, covering GPU prerequisites, REST API usage for file uploads, text/video embeddings, live RTSP streams, Redis/Kafka/OTel integration, and troubleshooting.
- ▌ Jetson Speculative Decoding · nvidia bundleReduce per-token latency on Jetson vLLM servers by appending speculative decoding configuration, with guidance on when to enable and how to benchmark the improvement.
- ▌ Tao Train Depth Anything V2 · nvidia bundleTrain, evaluate, export, and run inference for monocular depth estimation models using Metric Depth Anything v2 or Relative Depth Anything architectures via the TAO toolkit.
- ▌ Tao Train Foundation Stereo · nvidia bundleTrains, evaluates, exports, and runs inference on FoundationStereo models for stereo depth estimation and 3D reconstruction from stereo image pairs.
- ▌ Tao Train Mask Auto Encoder · nvidia bundleTrain, evaluate, export, and run inference for Masked Auto-Encoder (MAE) models for self-supervised pretraining and fine-tuning of visual representations.
- ▌ Tao Validate Dataset Format · nvidia bundleValidates NVIDIA TAO DAFT datasets for structure, schema, and cross-reference errors using the `tao-daft validate` CLI tool.
- ▌ Vss Deploy Dense Captioning · nvidia bundleDeploy a standalone RT-VLM dense-captioning microservice and exercise its REST API endpoints for file upload, caption generation, streaming, chat completions, and Kafka integration.
- ▌ Vss Manage Video Io Storage · nvidia bundleManage VIOS and NvStreamer REST API operations for video input/output and storage, including sensors, streams, snapshots, clips, and recordings.
- ▌ Mcore Linting And Formatting · nvidia bundleLint and format Python code for Megatron-LM using ruff, black, isort, pylint, and mypy, with commands for autoformatting and import ordering.
- ▌ Tao Generate Image Grounding · nvidia bundleGenerates phrase-grounded bounding box annotations from image-caption pairs using a VLM, producing cleaned captions, referring expressions, and pixel-space bounding boxes.
- ▌ Tao Run Automl Deft Pipeline · nvidia bundleRuns a three-phase AOI training pipeline: AutoML HPO baseline, DEFT iterative data improvement, and AutoML refinement on the augmented dataset.
- ▌ Tao Train Action Recognition · nvidia bundleTrain, evaluate, export, and run inference on TAO action-recognition models for classifying temporal actions in video clips using RGB, optical flow, or joint input.
- ▌ Tao Train Optical Inspection · nvidia bundleTrains, evaluates, exports, and runs inference for Siamese-network-based optical inspection models to detect manufacturing defects and quality issues in image pairs.
- ▌ Tilegym Adding Cutile Kernel · nvidia bundleAdd a new cuTile GPU kernel operator to TileGym, covering dispatch registration, backend implementation, exports, tests, and benchmarks.
- ▌ Vss Setup Behavior Analytics · nvidia bundleDeploy the behavior-analytics service standalone with a chosen entrypoint, config source, and optional calibration, without the full warehouse stack.
- ▌ Nemo Mbridge Multi Node Slurm · nvidia bundleConvert single-node PyTorch distributed scripts into multi-node Slurm sbatch jobs and debug common multi-node failures, covering srun-native and torch.distributed approaches, container setup, NCCL timeouts, and interactive allocation.
- ▌ Nemo Mbridge Perf Cuda Graphs · nvidia bundleValidate and use CUDA graph capture in Megatron Bridge, including local full-iteration graphs and Transformer Engine scoped graphs for attention, MLP, and MoE modules.
- ▌ Nv Generate Mr Brain Finetune · nvidia bundleFinetunes the NV-Generate-CTMR MR-brain diffusion UNet from user-supplied NIfTI training volumes using a wrapper that stages configs and delegates to upstream scripts.
- ▌ Tao Train Mask Grounding Dino · nvidia bundleTrains, evaluates, exports, quantizes, and runs inference for a Mask Grounding DINO model for open-set instance segmentation guided by text prompts.
- ▌ Tao Train Pose Classification · nvidia bundleTrain, evaluate, export, and run inference for pose classification models using ST-GCN on skeleton keypoint sequences.
- ▌ Vss Setup Video Analytics API · nvidia bundleDeploys the vss-video-analytics-api REST service standalone with config, data-log bind, and optional Elasticsearch/Kafka connectivity.
- ▌ Cupynumeric Parallel Data Load · nvidia bundleLoad sharded datasets (npy, Parquet, HDF5, raw binary) into distributed cuPyNumeric arrays using manual partitioning and Legate task launches.
- ▌ Deepstream Import Vision Model · nvidia bundleImport object detection models from HuggingFace or NVIDIA NGC into a DeepStream pipeline with automated ONNX download, TensorRT engine build, custom parser, multi-stream benchmark, and PDF report generation.
- ▌ Nemo Automodel Launcher Config · nvidia bundleConfigure NeMo AutoModel job launches for interactive runs, Slurm clusters, and SkyPilot cloud execution.
- ▌ Tao Finetune Huggingface Model · nvidia bundleFine-tune HuggingFace CV, VLM, or LLM models on local NVIDIA GPUs using an NGC PyTorch container, with support for full or LoRA training, dataset handling, and optional model push to the Hub.
- ▌ Tao Train Image Classification · nvidia bundleTrain, evaluate, distill, quantize, export, and run inference for PyTorch-based TAO image classification models with support for multiple backbones.
- ▌ Vss Generate Video Calibration · nvidia bundleRuns AutoMagicCalib calibration on local MP4s, RTSP streams, or a bundled sample dataset, and deploys the AMC microservice when needed.
- ▌ Cupynumeric Migration Readiness · nvidia bundleAssesses NumPy code for cuPyNumeric migration readiness by analyzing source code against an API support manifest and GPU-scaling idioms, producing a structured verdict with per-finding reasoning.
- ▌ Nemo Automodel Model Onboarding · nvidia bundleGuides implementation of new model architectures in NeMo AutoModel through five phases: discovery, implementation, registration, validation, and testing.
- ▌ Nemo Mbridge Perf Megatron Fsdp · nvidia bundleEnables Megatron Fully Sharded Data Parallel in Megatron-Bridge with configuration overrides, code anchors, pitfalls, and verification steps.
- ▌ Nemo Mbridge Perf Memory Tuning · nvidia bundleReduces peak GPU memory in Megatron Bridge training by applying expandable segments, parallelism resizing, activation recompute, and CPU offloading constraints.
- ▌ Nemo Mbridge Recipe Recommender · nvidia bundleIndexes Megatron Bridge recipes and recommends the best starting config based on model, GPU count, and training goal.
- ▌ Digital Health Clinical Asr Eval · nvidia bundleScore a clinical ASR manifest against a chosen NIM, produce a five-section KER leaderboard, and route the user via a post-eval decision tree.
- ▌ Nemo Mbridge Mlm Bridge Training · nvidia bundleRun Megatron-LM (MLM) and Megatron Bridge training with mock or real data, covering correlation testing, available recipes, and multi-GPU examples.
- ▌ Nemo Mbridge Perf Cpu Offloading · nvidia bundleConfigure and validate CPU offloading for Megatron Bridge training, including activation offloading and optimizer state offloading with HybridDeviceOptimizer.
- ▌ Omniverse Usd Performance Tuning · nvidia bundleDiagnose and optimize slow-loading, high-memory, or low-FPS USD scenes using a structured workflow with profiling, validation, and mutation phases.
- ▌ Tao Train Fast Foundation Stereo · nvidia bundleTrains, evaluates, exports, and runs inference for FastFoundationStereo (FFS) stereo depth estimation models, a distilled variant of FoundationStereo with lower latency.
- ▌ Vss Deploy Detection Tracking 2d · nvidia bundleDeploy, debug, and operate the RTVI-CV 2D detection/tracking microservice and call its REST API for stream management, health checks, and metrics.
- ▌ Vss Deploy Detection Tracking 3d · nvidia bundleDeploy and operate the RTVI-CV-3D microservice for multi-camera 3D detection and tracking, supporting sample datasets, custom videos, and RTSP streams.
- ▌ Digital Health Clinical Asr Build · nvidia bundleCurates clinical-specialty term lists, generates IPA-tagged synthetic audio via TTS, and produces NeMo-format manifests for ASR benchmark evaluation.
- ▌ Digital Health Clinical Asr Setup · nvidia bundleBootstraps a clinical ASR evaluation environment by verifying NVIDIA_API_KEY, installing Python dependencies, and running a smoke test against hosted TTS/ASR services.
- ▌ Nemo Automodel Recipe Development · nvidia bundleCreate and modify NeMo AutoModel training and evaluation recipes, including YAML structure, builders, and execution flow.
- ▌ Physical AI Neural Reconstruction · nvidia bundleRoutes NuRec/Neural Reconstruction requests to the correct upstream NVIDIA skill (datasets, conversion, training, rendering, object harvesting, frame cleanup).
- ▌ Tao Analyze Gaps Visual Changenet · nvidia bundleIdentifies the weakest samples per ground-truth label in NVIDIA TAO VCN Classify experiments by running a Docker container that performs threshold sweep, weakness scoring, and per-lighting expansion, then surfaces top-K weak samples for downstream augmentation or relabeling.
- ▌ Cuopt Numerical Optimization API C · nvidia bundleSolve LP, MILP, and QP problems using the cuOpt C API with a consistent build pattern and core calls.
- ▌ Nemo Mbridge Perf Moe Comm Overlap · nvidia bundleOptimizes MoE expert-parallel communication overlap in Megatron Bridge, covering dispatch/combine overlap, flex dispatcher backends, and expert wgrad scheduling.
- ▌ Nemo Mbridge Perf Moe Long Context · nvidia bundleProvides guidance for training Mixture-of-Experts models with long context windows, covering context parallelism sizing, selective recomputation, dispatcher choices, and practical patterns from recent experiments.
- ▌ Nemo Mbridge Perf Moe Vlm Training · nvidia bundleProvides practical guidance for training Mixture-of-Experts Vision-Language Models in Megatron Bridge, comparing FSDP and 3D-parallel approaches with lessons from recent multimodal experiments.
- ▌ Jetson Customize Uphy · nvidia bundleConfigure Jetson UPHY lane allocation on Orin/Thor custom carriers by editing ODMDATA tokens and dispatching per-controller kernel-DT overlay skills.
- ▌ Jetson Print Bsp Info · nvidia bundleInspects a Jetson Linux_for_Tegra BSP tree on the host PC and prints a concise summary including L4T version, board configs, and rootfs state.
- ▌ Jetson Validate Image · nvidia bundleRun static BSP checks and on-target smoke/regression tests on a flashed NVIDIA Jetson device to validate a customized BSP image.
- ▌ Nemo Evaluator Plugin · nvidia bundleRun evaluation tasks against a NeMo Platform server using the Evaluator plugin CLI and Python SDK.
- ▌ Nemo Rl Auto Research · nvidia bundleGuides agents through the full lifecycle of NeMo-RL experiments: understanding recipes, launching reproducible runs, analyzing results, and preserving human oversight with git and TSV logs.
- ▌ Tao List Capabilities · nvidia bundleLists TAO Skill Bank capabilities, models, and AutoML support by running packaged scripts.
- ▌ Tao Run On Kubernetes · nvidia bundleSubmits TAO container jobs as single-pod Kubernetes Jobs with NVIDIA GPU scheduling on EKS, GKE, AKS, or on-prem clusters.
- ▌ Tao Train Mask2former · nvidia bundleTrain, evaluate, export, quantize, and run inference on Mask2Former models for panoptic, instance, and semantic segmentation using NVIDIA TAO.
- ▌ Tao Train Single Step · nvidia bundleFine-tune a TAO model with standard supervised training, evaluation, and export, with AutoML bypass and platform-specific credential intake.
- ▌ Tilegym Cutile Python · nvidia bundleWrite high-performance GPU kernels using cuTile's tile-based programming model with validation and optimization, including deep agent orchestration for complex multi-kernel tasks.
- ▌ Dicom Metadata Extract · nvidia bundleExtracts selected metadata from a DICOM file and flags standard-tag PHI presence. Not for anonymization or clinical use.
- ▌ Dicom Series Preflight · nvidia bundleScans a DICOM series folder to extract header metadata and produce a preflight verdict without decoding pixel data.
- ▌ Dicom Series To Volume · nvidia bundleConverts a single CT DICOM series folder into a Hounsfield Unit NIfTI volume with affine metadata.
- ▌ Holoscan Install Conda · nvidia bundleInstall Holoscan SDK v4.3+ via Conda in a CUDA 13 environment, including Python bindings and C++ development headers.
- ▌ Holoscan Install Wheel · nvidia bundleInstall the Holoscan SDK Python wheel via pip into a virtual environment and verify with example scripts.
- ▌ Jetson Optimize Memory · nvidia bundleReclaim DRAM on NVIDIA Jetson devices by disabling unused display, camera, and DMA subsystems across MB1 BCT, MB2 BCT, kernel reserved-memory, and SWIOTLB layers for headless or no-camera deployments.
- ▌ Nemo Rl Brev Etiquette · nvidia bundleProvides storage and environment conventions for NeMo-RL agents on Brev instances, ensuring large experiment outputs go to /ephemeral and secrets are loaded from .env.
- ▌ Nemo Rl Session Memory · nvidia bundleMaintain durable session memory across agent disconnects by writing structured checkpoints to the repo's session directory, enabling context recovery.
- ▌ Nv Segment Ct Finetune · nvidia bundleFine-tune NV-Segment-CT VISTA3D on CT NIfTI labels for smoke testing or dataset adaptation, wrapping the upstream MONAI bundle entrypoint.
- ▌ Tao Train Nvpanoptix3d · nvidia bundleTrains, evaluates, exports, and runs inference for NVPanoptix3D models that perform panoptic 3D scene reconstruction from posed RGB images, producing 3D panoptic segmentation with occupancy completion.
- ▌ Tao Train Pointpillars · nvidia bundleTrain, evaluate, export, prune, and run inference for PointPillars 3D object detection models from LiDAR point clouds using NVIDIA TAO.
- ▌ Cuopt Server API Python · nvidia bundleDeploy a cuOpt REST server and solve routing optimization problems using Python or curl.
- ▌ Earth2studio Data Fetch · nvidia bundleGuides users through downloading weather/climate data via Earth2Studio data source APIs, verifying variable support via the lexicon system, and generating a working Python fetch script.
- ▌ Holoscan Install Debian · nvidia bundleInstall the Holoscan SDK C++ runtime and headers on Ubuntu using NVIDIA's apt repository, with automatic CUDA variant detection and verification via bundled examples.
- ▌ Holoscan Install Source · nvidia bundleBuild the Holoscan SDK from source using its in-tree Docker-based build script, producing a local install tree for CMake-based applications.