Verified skills

Build the Holoscan SDK from source using its in-tree Docker-based build script, producing a local install tree for CMake-based applications.

Enable MIPI/GMSL camera sensors on a Jetson Thor or Orin custom carrier by rendering a kernel-DT overlay from the in-tree sensor DTSI.

Lock, cap, or customize CPU, GPU, and EMC clock behavior on NVIDIA Jetson devices by editing BPMP DTB and nvpower.sh before flashing.

Configure fault tolerance, straggler detection, preemption, in-process restart, and re-run state machine for Megatron Bridge training jobs.

Run TAO SDK jobs as Docker containers on a local or remote Docker daemon with NVIDIA GPU support, including preflight checks and credential handling.

Solve vehicle routing problems (TSP, VRP, PDP) using NVIDIA cuOpt's Python API with cost matrices, time windows, capacity constraints, and pickup-delivery pairs.

Captures a baseline snapshot of a Jetson device's module model, L4T version, kernel, OS version, and power mode for performance testing or verification.

Extract false-positive and false-negative gaps from VLM binary-classification-question predictions by comparing model responses against ground truth, producing a structured JSONL file and summary report for downstream root-cause analysis.

Trains, evaluates, exports, quantizes, and runs inference for a Grounding DINO model that detects objects described by text prompts without a fixed class vocabulary.

Defines vehicle routing problem types (TSP, VRP, PDP) and the data requirements needed to formulate them, without covering any API or interface details.

Validates that a Dynamo deployment's NIXL/UCX/NCCL interconnect is ready for disaggregated serving over RDMA/NVLink. Use after deploying a disagg or multi-node recipe to confirm KV transport is correct, or use troubleshoot for already-failed pods.

Recommends an inference runtime and memory-related launch flags for LLM/VLM workloads on NVIDIA Jetson devices, based on a live memory audit snapshot.

Build synthetic datasets and data generation pipelines using the Data Designer library.

Generates custom safety policies for NVIDIA Nemotron content-safety guardrails, producing a Markdown policy, JSON taxonomy, and inference prompts from rough user input.

Coordinate the end-to-end pipeline from CAD or source assets to SimReady USD assets, including conversion, material and physics assignment, validation, conformance, and optional packaging.

Performs deep root cause analysis on NVIDIA TAO Visual ChangeNet classification experiments, using image-evidence-driven investigation to diagnose model failures and produce actionable reports.

Fine-tune, evaluate, run inference, and export Cosmos-Embed1 video-text embedding models for tasks like text-to-video retrieval and semantic deduplication.

Start, query, and stop a TAO inference microservice for a specific network architecture by delegating container execution to the appropriate platform skill.

Checks and installs NVIDIA driver, CUDA Toolkit, and NVIDIA Container Toolkit for GPU-accelerated Docker and Kubernetes hosts. Supports multiple Linux distributions with automated install and read-only check modes.

Trains, evaluates, and runs inference for Mask Auto-Label (MAL) weakly-supervised segmentation models using ViT-MAE backbones with minimal point or box annotations.

Adds autotuning to CuTile kernels using the exhaustive_search API with a tune-once/cache/direct-launch pattern, covering occupancy-only and complex tile-size search spaces.

Generates 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.

Accelerate pandas workflows with GPU DataFrames using cuDF and dask-cuDF for ETL, joins, groupby, and large-scale data processing.

Pull and verify the official Holoscan SDK container from NGC, selecting the correct CUDA/arch tag for the host GPU and validating with bundled Python and C++ examples.

Plan, debug, tune, evaluate, export, or deploy public Nemotron embedding and reranking retrieval recipes using the current checkout.

Converts NVIDIA TAO DAFT datasets between supported formats using the `tao-daft convert` CLI.

Fine-tune Cosmos Reason video QA models using supervised fine-tuning with FSDP parallelism, including dataset preparation, spec construction, and AutoML support.

Integrate a HuggingFace computer vision model into the NVIDIA TAO Toolkit ecosystem, covering the full pipeline from prerequisites to container testing.

Trains, 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.

Deploy 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.

Reduce per-token latency on Jetson vLLM servers by appending speculative decoding configuration, with guidance on when to enable and how to benchmark the improvement.

Train, evaluate, export, and run inference for monocular depth estimation models using Metric Depth Anything v2 or Relative Depth Anything architectures via the TAO toolkit.

Trains, evaluates, exports, and runs inference on FoundationStereo models for stereo depth estimation and 3D reconstruction from stereo image pairs.

Train, evaluate, export, and run inference for Masked Auto-Encoder (MAE) models for self-supervised pretraining and fine-tuning of visual representations.

Deploy a standalone RT-VLM dense-captioning microservice and exercise its REST API endpoints for file upload, caption generation, streaming, chat completions, and Kafka integration.

Manage VIOS and NvStreamer REST API operations for video input/output and storage, including sensors, streams, snapshots, clips, and recordings.