Total 50,527 skills, AI & Machine Learning has 8482 skills
Showing 12 of 8482 skills
After solving a non-trivial problem, detect generalizable learnings and propose skill updates so future interactions benefit automatically. Always active — applies to every interaction.
Give your AI agents capabilities through tools (function calling). Helps you identify what your AI needs to do, create tool definitions, and attach them to AI Config variations.
End-to-end orchestration for non-trivial software feature development. Use this skill whenever the user asks to implement a PR-sized feature, break down a plan, have subagents review a plan, run a plan-review-development-acceptance loop, coordinate multiple review perspectives, produce an acceptance report, or generate an HTML PR summary. Prefer this skill for multi-step code changes even if the user only says "build this feature" and the task is not a tiny one-file edit.
Guide for adding support for new LLM or VLM models in Megatron-Bridge. Covers bridge, provider, recipe, tests, docs, and examples.
cuOpt REST server — what it does and how requests flow. Domain concepts; no deploy or client code.
Code instrumentation for timing workloads. Two scenarios: (1) Training loop — inject manual timing to report per-iteration latency, throughput (samples/sec), and data load time. (2) Standalone kernel/op — write CUDA event timing code with warmup, per-iteration statistics, and anti-pattern avoidance. Also covers NVTX annotation for labeling profiler timelines. NOT for: running or analyzing profiler tools (nsys, ncu, Nsight Systems, Nsight Compute), writing kernels (Triton, CuTe, CUDA), applying optimizations (CUDA Graphs, gradient checkpointing, fusion), or interpreting roofline/SOL% metrics. Triggers: "measure throughput", "benchmark this function", "time my training loop", "samples per second", "NVTX annotate", "instrument my dataloader", "data load time", "kernel timing", "how do I time".
Produce video analysis reports by discovering the deployed VSS agent, querying POST /generate for a timestamped captioned summary of the clip, then formatting the agent reply as the standard Video Analysis Report markdown.
ONLY for OpenAI Triton (@triton.jit) kernel development. NEVER use for CUDA C++ kernels, TileIR, or profiling tools (ncu, nsys). The user's request must involve Triton explicitly. Covers Triton-specific patterns: fused elementwise, reductions (softmax, LayerNorm, RMSNorm), tiled GEMM with triton.autotune, and flash attention. Workflow: design, write, verify (with fast-path for explicit requests).
Call the vss agent to run video understanding on video to answer a text question. Use when the user asks about video content, or about visual details that cannot be answered from conversation history, search hits, or metadata alone.
Brev instance operating guidance for NeMo-RL agents working in /home/ubuntu/RL with limited workspace disk, a larger /ephemeral volume, and optional /home/ubuntu/RL/.env secrets. Use when running auto-research campaigns, experiments, training jobs, model or dataset downloads, shared cache-heavy commands, log-producing runs, checkpoint generation, W&B or Hugging Face authenticated workflows, or any workflow that may create large files on Brev.
Use when the user asks to define a goal, create a Goal Contract, or clarify a concrete task's goal, scope, success criteria, evidence, or guardrails before planning or execution.
ADBPG Knowledge Base Management: Create knowledge bases, upload documents, search, Q&A. Triggers: "knowledge base", "document library", "document upload", "knowledge search", "RAG", "Q&A", "embedding", "ADBPG", "AnalyticDB PostgreSQL"