skillsNVIDIA/skills
An official NVIDIA engineering skills repository for accelerated computing, TensorRT optimization, and embodied AI.…
No ratings yet
Rate this skill
Stars3.7k
Forks457
Watchers3.7k
Open Issues16
InstallOne-line install
UnconfirmedWhether this repository follows the Agent Skills spec (that is, whether it contains a SKILL.md) has not been verified yet. Until that is settled, clone it and use what you need:
git clone https://github.com/NVIDIA/skills.git
OverviewOverview
Key Features
An official NVIDIA engineering skills repository for accelerated computing, TensorRT optimization, and embodied AI. Key capabilities include: 1. full-stack GPU kernel optimization (demonstrates CUDA, CuDNN, and TensorRT techniques to slash LLM latency and VRAM footprint); 2. physical simulation and robotics environments (interfaces with Isaac Sim and Omniverse for synthetic data generation and RL training); 3. multi-GPU distributed orchestration (implements Megatron-LM tensor and pipeline parallelism across high-speed NVLink fabrics).
Use Cases
High-Concurrency Performance Tuning, Deep Learning Deployment, System Architecture Design.
Target Roles
AI / Algorithm Engineer, Backend Engineer.
TopicsTopics
RatingUser ratings
Nobody has rated this skill yet. Yours would be the first on the site — your vote becomes the initial average.
RepositoryRepository
| Skill ID | 304 |
|---|---|
| Data source | GitHub |
| Repository | NVIDIA/skills |
| Category | AI Agent Architecture & RAG / Agent Orchestration & Workflow |
| Primary language | Python |
| License | Apache License 2.0 |
| Default branch | main |
| GitHub repository ID | 1167034425 |
| Node ID | R_kgDORY-IOQ |
| Rank | #336 (position within the source ranking) |
| Site-wide popularity rank | #336 (computed live from Stars, across all 5833 entries) |
GrowthStar / Fork growth
One snapshot per day, tracking how Stars and Forks changed
- Now3,719
- Previous point · 10-25 3,719
- vs previous point ±0
- 4 data points 36 days
- Now457
- Previous point · 10-25 457
- vs previous point ±0
- 4 data points 36 days
TimelineTimeline
Repository created2026-02-25 21:54:058 months ago
Data last updated2026-11-05 13:19:3613 days ago
Last commit2026-10-24 16:12:2225 days ago
CommitLatest commit
Committed by mosheabr. Update Category: Dependency Upgrades & Maintenance. Commit details: chore(metadata): regenerate metadata.json, skills.sh.json, benchmarks.json, and versions.json. This update refreshes dependencies and build configuration.