Claude Skills for a Data Scientist / ML — Page 6
TAO OneFormer Universal Segmentation
Train and deploy one task-conditioned model for panoptic, instance, and semantic image segmentation.
TileGym cuTile Kernel Guide
A step-by-step guide for registering, implementing, testing, and benchmarking a new cuTile GPU kernel in TileGym.
Data Designer Synthetic Data Skill
Build synthetic datasets and data-generation pipelines from a natural-language specification.
DICOM Metadata Extractor
Extract selected metadata from one DICOM file and flag PHI in standard tags.
Clinical ASR Benchmark Builder
Build pronunciation-aware NeMo evaluation benchmarks for clinical speech recognition.
DOCA GPI GPU-Initiated RDMA Skill
Helps CUDA kernels drive RDMA queues directly from GPU memory without host-CPU mediation.
GPUNetIO RDMA Write Latency Benchmark
Guides reproducible measurement of CUDA-kernel-initiated GPUNetIO RDMA WRITE latency and tail behavior.
DOCA SHA OpenSSL Offload Engine
Adds DOCA SHA hardware acceleration to existing OpenSSL pipelines without rewriting the application around the doca-sha C API.
DOCA UROM Host-Side Offload Development
Guides HPC, UCX, and MPI developers in offloading remote memory operations to a BlueField DPU.
Holoscan Conda Installer
Install and validate Holoscan SDK on Linux x86_64 with a CUDA 13 driver using Conda.
Holoscan Native Ubuntu Installer
Install and verify the Holoscan SDK C++ environment on Ubuntu from NVIDIA’s apt repository.
Jetson Document Binder
Registers staged Jetson reference documents in the active target profile.
Jetson Memory Optimizer
Reclaim unused DRAM on headless or no-camera Jetson deployments.
Megatron-Core Testing Guide
Design, run, and reproduce distributed Megatron-Core tests with CI-aligned workflows.
MoE Dispatcher Selection Guide
Choose alltoall, DeepEP, or HybridEP from the hardware, expert-parallel degree, and optimization stage.
MoE Hardware Configuration Reference
Plan MoE training layouts and throughput expectations across NVIDIA GPU platforms.
NV-Reason-CXR Chest X-ray Reasoning Test
Runs reproducible command-shape and live inference smoke tests for chest X-ray reasoning.
Omniverse USD Performance Tuning
Diagnose USD scene performance problems and coordinate evidence-based optimization from runtime setup through reporting.
TAO Image Grounding
Convert image-caption pairs into phrase-grounded annotations with pixel-space bounding boxes.
TAO FoundationStereo Stereo Depth
Generate disparity maps from stereo images and manage FoundationStereo training, evaluation, and deployment for 3D reconstruction.
TAO PointPillars 3D Detection
Guides TAO workflows for training, evaluating, and deploying LiDAR-based PointPillars detectors.
TAO SegFormer Semantic Segmentation
Guidance for training, evaluating, exporting, and deploying TAO SegFormer models.
TAO Standard Training Workflow
Run a controlled TAO train, evaluate, and export workflow on labeled data.
CuTile Autotuning Guide
Design, implement, and validate low-overhead autotuning for CuTile GPU kernels.