Add unsigned integer (uint) type support to PyTorch operators by updating AT_DISPATCH macros. Use when adding support for uint16, uint32, uint64 types to operators, kernels, or when user mentions enabling unsigned types, barebones unsigned types, or uint support.
Hardware-agnostic quantum ML framework with automatic differentiation. Use when training quantum circuits via gradients, building hybrid quantum-classical models, or needing device portability across IBM/Google/Rigetti/IonQ. Best for variational algorithms (VQE, QAOA), quantum neural networks, and integration with PyTorch/JAX/TensorFlow. For hardware-specific optimizations use qiskit (IBM) or cirq (Google); for open quantum systems use qutip.
ToolUniverse workflow — Variant Interpretation
Deep learning framework (PyTorch Lightning). Organize PyTorch code into LightningModules, configure Trainers for multi-GPU/TPU, implement data pipelines, callbacks, logging (W&B, TensorBoard), distributed training (DDP, FSDP, DeepSpeed), for scalable neural network training.
Expert GPU optimization for modern consumer GPUs (8-24GB VRAM). Use this skill when you need to optimize GPU training, speed up CUDA code, reduce OOM errors, tune XGBoost for GPU, migrate NumPy to CuPy, make a model faster, manage GPU memory, optimize VRAM usage, or benchmark PyTorch. Covers mixed precision, gradient checkpointing, XGBoost GPU acceleration, CuPy/cuDF migration, vectorization, torch.compile, and diagnostics. NVIDIA GPUs only. PyTorch, XGBoost, and RAPIDS frameworks.
Performance optimization guidelines for Splitrail. Use when optimizing parsing, reducing memory usage, or improving throughput.
Infers backend technologies including servers, languages, frameworks, databases, and CMS
Guide and execute the migration of asynchronous code from RxJava to Kotlin Coroutines and Flow. Use this skill when a user asks to convert RxJava (Observables, Singles, Completables, Subjects) to Coroutines (suspend functions, Flows, StateFlows).
ComfyUI advanced node patterns - MatchType, Autogrow, DynamicCombo, DynamicSlot, node expansion, MultiType, wildcard inputs. Use when building complex nodes with dynamic inputs, type matching, or node expansion.
Compile and flash ESPHome firmware locally to a USB-connected ESP32 device using Docker. Use when the user says "deploy locally", "dploy over usb", "build and flash", "upload firmware", "flash over USB", "test locally", or wants to test component changes on a physical device.
Port FlyDSL GPU kernels from raw buffer_ops (create_buffer_resource, buffer_load, buffer_store with manual byte-offset arithmetic) to the layout API (make_buffer_tensor + logical_divide + copy_atom_call with BufferCopy atoms). Use when a kernel uses raw buffer_ops and should be migrated to the higher-level layout algebra for consistency and readability. Usage: /port-to-layout-api <kernel_file>
Refactoring MCP Tools for Better LLM Integration and Usability
Apply prefetch optimization to FlyDSL kernel loops: pre-load the first iteration's data before the loop, issue async loads for the next iteration inside the loop body, and swap buffers at the loop tail via scf.for loop-carried values. This overlaps data load latency with compute instructions. Use when a kernel has a loop where buffer_load feeds into MFMA/compute and load latency is exposed. Usage: /prefetch-data-load
Comprehensive guide to optimizing GEMM (General Matrix Multiply) kernels in FlyDSL on AMD CDNA GPUs. Covers tiling strategy, LDS ping-pong double-buffer, XOR bank-conflict swizzle, A/B data prefetch pipeline, 2-stage software pipelining, MFMA instruction scheduling (hot_loop_scheduler), epilogue strategies (direct store vs CShuffle), TFLOPS/bandwidth calculation, main-loop instruction count analysis, and bottleneck identification from ATT traces. Based on the production preshuffle_gemm kernel. Usage: /gemm-optimization
Use rsbkb for binary data manipulation, CLI tools: hex unhex urlenc urldec crc16 crc32 crc b64 d64 bofpattoff bofpatt xor entropy slice bgrep findso tsdec tsenc deflate inflate base escape unescape
Helps write CUDA and HIP kernels using torch.utils.cpp_extension.load_inline(). Use when users want to write native GPU code (CUDA/HIP) inside a Python submission file.
Helps prepare and submit popcorn-cli GPU Mode solutions. Use when users ask to set up a project, create a submission template, or run/register submissions.
LLM integration patterns for function calling, streaming responses, local inference with Ollama, and fine-tuning customization. Use when implementing tool use, SSE streaming, local model deployment, LoRA/QLoRA fine-tuning, or multi-provider LLM APIs.
Translate a threshold concept from disciplinary language into developmentally appropriate Know/Understand/Do for a specific band. Use when building programme-level KUD frameworks from threshold concepts.
Translate an external competency framework like DigComp, GreenComp, or ISTE into classroom-ready activities. Use when implementing framework standards in specific teaching contexts.
Design a structured goal-setting protocol using SMART or implementation-intention frameworks for students. Use when launching units, projects, or developing student self-direction habits.
Manages persistent research memory across ideation and experimentation cycles. Maintains two stores: Ideation Memory M_I (feasible/unsuccessful directions) and Experimentation Memory M_E (reusable strategies for data processing, model training, architecture, debugging). Three evolution mechanisms: IDE (after idea-tournament), IVE (after experiment failure — classifies failures as implementation vs fundamental), ESE (after experiment success — extracts reusable strategies). Use when: updating memory after completing idea tournaments or experiment pipelines, classifying why a method failed (implementation vs fundamental failure), starting a new research cycle needing prior knowledge, user mentions 'update memory', 'classify failure', 'what worked before', 'research history', 'evolution'. Do NOT use for running experiments (use experiment-pipeline), debugging experiment code (use experiment-craft), or generating ideas (use idea-tournament).
