AI-Hypercomputer

Use this skill when asked to generate or update release notes in docs/release_notes.md with major changes, new features, architecture updates, performance improvements, bug fixes, and deprecations.

Performs new LLM/VLM model bring-up in MaxText on TPU VMs. Use when onboarding a new HuggingFace model (e.g., Qwen, Llama, Gemma variants) to MaxText. This includes analyzing the HuggingFace source code, implementing decoder layers in JAX/Flax, writing…

Prepares, validates, uploads, and shepherds pull requests to the AI-Hypercomputer/maxtext GitHub repository through CI and code-owner review. Use when formatting or linting MaxText code, writing MaxText commit messages or PR descriptions, stacking or rebasing…

Orchestrates Pallas TPU kernel development and optimization. Converts JAX, PyTorch or CUDA into a Pallas TPU kernel and runs a self-refinement loop (plan -> implement -> compile -> test -> autotune -> profile) by dispatching the maxkernel-worker subagent once…