Verified skills

Submit and monitor GPU training jobs on Brev, SLURM, Docker, or Kubernetes using the TAO Execution SDK, with job handles, S3 I/O wrapping, and multi-node distributed training.

Trains, evaluates, exports, prunes, quantizes, retrains, and runs inference for OCDNet scene text detection models using TAO, detecting arbitrary-oriented text regions in natural images.

Trains, evaluates, exports, prunes, quantizes, retrains, and runs inference for TAO OCRNet models for scene text recognition from cropped text-region images, supporting CTC and attention-based decoders.

Train, evaluate, distill, quantize, export, and run inference for RT-DETR object detection models using NVIDIA TAO.

Write, review, and migrate code using NVIDIA DALI's imperative dynamic-mode API for efficient data loading and preprocessing.

Captures a read-only health snapshot from a Jetson device, reporting identity, memory, GPU, thermal, power, storage, services, and top processes.

Extract Jetson Linux BSP and sample-rootfs tarballs, run apply_binaries.sh with the correct GPU stack flag, and record the image metadata in the active target profile.

Switch the active Jetson target-platform pointer to an existing profile YAML. Use before customize/build/flash to change target; not for authoring profiles — use jetson-init-target instead.

Fine-tune and deploy CLIP vision-language models for zero-shot classification, image-text retrieval, and embedding extraction with ONNX and TensorRT support.

Submit TAO jobs to Lepton managed GPU compute on DGX Cloud, with run/status/cancel interface and multi-node distributed training support.

Operate the VSS alert pipeline for real-time monitoring, Alert-Bridge subscriptions, Slack notifications, incident queries, and camera onboarding.

Flash a promoted BSP image to a Jetson device in RCM mode using NVIDIA's flash.sh or l4t_initrd_flash.sh toolchain.

Generates a per-target knowledge-base markdown file by walking the BSP root and source tree, documenting image layout, source tree structure, and document references.

Bootstraps a BSP customization workspace for NVIDIA Jetson, including the Linux_for_Tegra overlay tracker, bsp_sources mono-tree, and Crosstool-NG toolchain.

Creates a new Jetson target-platform profile YAML by selecting a reference devkit and optional custom carrier, then updates the active target pointer.

Dispatches Jetson BSP customization by presenting a click-to-select setup questionnaire and passing prefilled answers to downstream setup skills.

Investigate a failing GitHub Actions job, extract the root cause, and file a well-structured bug issue against NVIDIA/Megatron-LM.

Launch distributed Megatron-LM training jobs on a SLURM cluster with a minimal sbatch skeleton, environment-variable setup for torch.distributed.run, CUDA_DEVICE_MAX_CONNECTIONS rules, container conventions, monitoring, and per-rank failure diagnosis.

Plan, configure, and chain Nemotron model customization steps into single-step or multi-step pipelines for curation, translation, fine-tuning, RL alignment, benchmarking, checkpoint conversion, optimization, and evaluation.

Trains vision transformers via self-distillation without labels for self-supervised visual representation learning, and supports export and inference of NVDINOv2 backbones.

Trains, evaluates, exports, quantizes, and runs inference for Sparse4D multi-camera temporal 3D object detection and tracking models using TAO.

Selects, configures, deploys, verifies, debugs, or tears down a VSS profile (base, search, lvs, warehouse, edge) for NVIDIA's video search and summarization stack.

Search archived video using natural language, ingest video files or RTSP streams, and manage ingested sources.

Install and verify cuPyNumeric for Python using conda or pip, including GPU usage checks.

Diagnose failed or unhealthy Dynamo deployments by collecting a read-only debug bundle, classifying failures, and providing step-by-step remediation guidance.

Rebuild kernel-side artifacts (DTBs, OOT modules, kernel Image) from source changes under bsp_sources/ and produce a manifest for staging into a BSP image.

Downloads NVIDIA Jetson Linux BSP artifacts (BSP tarball, sample rootfs, public_sources, x-tools, guides) for the active target. Used for Auto Setup; does not extract or edit profiles.

Measure Jetson DRAM and NvMap usage, capture before/after baselines, and verify memory reclamation with live audit data.

Guides AI coding assistants to the official NemoClaw documentation via MCP server or Markdown files for installation, configuration, operation, and troubleshooting.

Collects launch inputs and runs preflight checks before executing TAO workflows such as AutoML, training, evaluation, inference, export, TensorRT engine generation, or DEFT jobs on supported platforms.

Embeds target and source image parquets, then mines nearest-neighbour source images for augmentation in VCN AOI workflows.

Trains, evaluates, and runs inference for BEVFusion multi-sensor 3D object detection models that fuse LiDAR and camera data in bird's-eye-view space for autonomous driving.

Train, evaluate, export, quantize, and run inference for a TAO OneFormer model that performs panoptic, instance, and semantic segmentation using task-conditioned queries.

Trains, evaluates, exports, quantizes, and runs inference for SegFormer semantic segmentation models using NVIDIA TAO.

Queries video analytics incidents, alerts, metrics, and sensor data from Elasticsearch via the VA-MCP server.

Summarize recorded video clips using the LVS microservice with a VLM fallback, producing a narrative summary with timestamped events.