Loading...
Loading...
Found 34 Skills
Triton Inference Config - Auto-activating skill for ML Deployment. Triggers on: triton inference config, triton inference config Part of the ML Deployment skill category.
Model Registry Manager - Auto-activating skill for ML Deployment. Triggers on: model registry manager, model registry manager Part of the ML Deployment skill category.
Feature Store Connector - Auto-activating skill for ML Deployment. Triggers on: feature store connector, feature store connector Part of the ML Deployment skill category.
Model Export Helper - Auto-activating skill for ML Deployment. Triggers on: model export helper, model export helper Part of the ML Deployment skill category.
Azure Ml Deployer - Auto-activating skill for ML Deployment. Triggers on: azure ml deployer, azure ml deployer Part of the ML Deployment skill category.
Onnx Converter - Auto-activating skill for ML Deployment. Triggers on: onnx converter, onnx converter Part of the ML Deployment skill category.
Serve a model with MAX's `max serve` command: set up the environment (pixi or uv with the max-nightly conda channel / nightly wheel index), point the server at a Hugging Face repo or local checkpoint, target a custom architecture with `--custom-architectures`, and pick the right serve flags for the model. Use this whenever the user wants to run, launch, start, or host a model on MAX, bring up an OpenAI-compatible endpoint, serve a custom/ported architecture, debug a `max serve` startup failure, or figure out which serve flags (devices, quantization-encoding, max-length, task, trust-remote-code) a given model needs, even if they don't say "max serve" by name.
Use when a locally-served model misbehaves in an agent harness — wrong/empty tool calls, truncated output, context overflow, or slow generation. Diagnoses whether the fault is the model tier, the serving config, or the harness wiring, using the agentic repo's probe and model matrix.
This skill should be used when the user asks to optimize, export, serve, compress, or accelerate ML inference. PROACTIVELY activate for: (1) latency, throughput, p95/p99, batching, concurrency, KV cache, memory, or cost issues, (2) quantization INT8/INT4, GPTQ, AWQ, bitsandbytes, pruning, sparsity, distillation, (3) ONNX export, ONNX Runtime, TensorRT, TorchScript, torch.compile, XLA, OpenVINO, Core ML, TFLite, (4) Triton, TorchServe, TF Serving, BentoML, Seldon, KServe configuration, (5) edge deployment, CPU/GPU/TPU/Inferentia serving. Provides: hardware-aware inference optimization and safe benchmarking.
Debug inference clients that use an attached provider and its native endpoint, including hosted APIs and host-local Ollama, vLLM, SGLang, TRT-LLM, LM Studio, or NIM. Use for provider attachment, endpoint policy, credential substitution, topology, and migration from the removed managed inference endpoint. Trigger keywords - debug inference, managed inference endpoint, local inference, ollama, lm studio, vllm, sglang, trtllm, NIM, inference failing, model server unreachable, credential_endpoint_mismatch, host.openshell.internal.