Coding standards, frameworks, refactoring, and code review (15.8K skills)
Implement memory copies with DOCA DMA, which transfers data using the device rather than the host CPU. Get guidance on initialization, buffer limits, memory permissions, task completion, and diagnosing errors against the locally installed SDK.
Build and troubleshoot applications that move DEFLATE compression and decompression, or LZ4 decompression, to DOCA-supported hardware. Guidance covers device capabilities, buffer sizes, memory permissions, validation, and deciding when CPU processing is more suitable.
Guide DOCA Management programming for BlueField and ConnectX devices. Set up management contexts, query capabilities, change device settings, and investigate errors, using installed reference samples and checks before and after changes.
Build and troubleshoot custom plug-ins for workloads that doca-bench cannot measure with its built-in modes. Use the shipped CUDA example to guide shared-library construction, check loading and version compatibility, and verify a small test before larger runs.
Implement and troubleshoot storage recovery with DOCA Erasure Coding. The skill guides parity generation, lost-block recovery, and incremental parity updates using Reed-Solomon coding, while checking device capabilities, block limits, and memory permissions.
Build SHA hashing applications for BlueField or ConnectX accelerators with guidance on single-request versus chunked processing, supported algorithms, buffer sizes, memory permissions, and errors returned by the DOCA SHA API.
Load and configure a custom congestion control algorithm on a BlueField DPU from the host. The skill guides port attachment, capability checks, and troubleshooting when loading fails or the algorithm has no effect on RoCE traffic.
Modify command-line options in DOCA applications while preserving standard flags. This skill guides parameter registration, parser lifecycle, JSON configuration, and error diagnosis using the installed SDK and its shipped samples.
Develop messaging between a host and BlueField using DOCA Comch. Guides server and client setup, selection of simple messaging or higher-throughput data transfer, capability checks, and diagnosis of connection, permission, and API errors.
Develop DOCA Rivermax applications that receive timing-sensitive video, audio, and other data over IP networks. Get guidance on receive stream setup, device capability checks, and troubleshooting errors or missing receive events.
Build DOCA applications that send and receive network packets directly from CUDA code on an NVIDIA GPU. Get guidance on connecting queues, registering GPU memory, checking hardware capabilities, and diagnosing GPUNetIO setup and runtime errors.
Build custom emulated PCIe devices that a host recognizes as peripherals while BlueField DPU code handles their operation. Get guidance on DOCA sub-library selection, capability checks, setup within an installed environment, and debugging device visibility or driver binding.
Investigate DOCA build failures, runtime errors, and silent tools by checking each system layer. Get guidance on increasing log detail and collecting reproducible evidence for troubleshooting or a Developer Forum question.
Develop host-side applications that run work on BlueField's DPA, a data-path processor. Load compiled application images, launch kernels, observe completion, check supported features, and investigate errors involving hardware capabilities or DOCA and DPACC compatibility.
Connect existing DOCA Flow processing rules and resources to BlueField’s DPA processor. The skill guides export sequencing, operation queues, counter access, and entry or memory updates, and helps diagnose failures between host setup and device-side access.
Build and debug the shared foundations of DOCA applications on BlueField or ConnectX hardware. Get guidance on device capabilities, memory buffers, context lifecycles, task completion processing, and separate logging controls for the SDK and application.
Guides AES-GCM encryption and decryption development with DOCA on BlueField or ConnectX hardware. Covers supported key sizes, buffer permissions, published test vectors, and authentication tags, which detect altered data, including how to handle verification failures.
Start a cuOpt REST server and build Python or curl clients to submit routing and optimization problems. Follow result polling, translate Python API field names into REST fields, and diagnose rejected requests.
Customize NVIDIA Nemotron Voice Agent’s Generic Pipecat example for appointment scheduling, five-field patient intake, or a custom healthcare conversation. Build prompts and tool handlers without a separate backend, then validate conversation behavior, services, and voice interaction.
Modify NVIDIA cuOpt itself, covering C++/CUDA solver code, Python bindings, server endpoints, and CI. Follow project conventions for environment setup, builds, tests, debugging, and contributions, including the required sign-off for submitted changes.
Build video analytics applications with NVIDIA DeepStream and Python's pyservicemaker API. The skill guides video input, object detection, optional tracking, annotated output, and Kafka messaging, with configuration guidance and fixes for common pipeline errors.
Measure and tune a DeepStream video pipeline using Nsight Systems. The skill benchmarks inference batch sizes, derives configuration settings from measurements and GPU limits, and reports FPS, processing bottlenecks, and available timings for individual processing components.
Get guidance on installing CUDA-Q, writing quantum processing functions, and choosing CPU or GPU simulation targets. Also covers connecting to quantum processors, running a Bell-state example, and selecting approaches for execution across multiple devices.
Port quantum circuits from Qiskit or other frameworks into CUDA-Q while preserving the original algorithm. Guides gate conversion, measurement ordering, precision checks, and small test comparisons to produce code that runs without the source framework.