Data pipelines, feature stores, and embedding generation for AI/ML systems. Use when building RAG pipelines, ML feature serving, or data transformations. Covers feature stores (Feast, Tecton), embedding pipelines, chunking strategies, orchestration (Dagster, Prefect, Airflow), dbt transformations, data versioning (LakeFS), and experiment tracking (MLflow, W&B).
Optimizing vector embeddings for RAG systems through model selection, chunking strategies, caching, and performance tuning. Use when building semantic search, RAG pipelines, or document retrieval systems that require cost-effective, high-quality embeddings.
Strategic guidance for operationalizing machine learning models from experimentation to production. Covers experiment tracking (MLflow, Weights & Biases), model registry and versioning, feature stores (Feast, Tecton), model serving patterns (Seldon, KServe, BentoML), ML pipeline orchestration (Kubeflow, Airflow), and model monitoring (drift detection, observability). Use when designing ML infrastructure, selecting MLOps platforms, implementing continuous training pipelines, or establishing model governance.
LLM and ML model deployment for inference. Use when serving models in production, building AI APIs, or optimizing inference. Covers vLLM (LLM serving), TensorRT-LLM (GPU optimization), Ollama (local), BentoML (ML deployment), Triton (multi-model), LangChain (orchestration), LlamaIndex (RAG), and streaming patterns.
Build Mprove Model based on Malloy sources
Toyota 5 Whys methodology with multi-causal branching, evidence requirements, and validation techniques
Clojure language-specific patterns, data-first modeling, REPL-driven development, and spec
F# language-specific patterns, Railway-Oriented Programming, and Computation Expressions
Stateful property-based testing patterns, model-based testing, and anti-patterns
4-step system design framework with back-of-envelope estimation, scaling ladder, and common pitfalls
Preflight a tx via eth_estimateGas with optional revert reason.
Feature development with cross-module impact analysis. Covers trainer hierarchy, model adapters, reward pipeline, config system, sample dataclasses, and distributed training paths. Trigger: 'add feature', 'implement', 'refactor', 'reorganize', 'new capability'.
Complete workflow for adding a new RL training algorithm. Covers paradigm selection, TrainingArguments subclass, trainer implementation, registry, example config, and verification. Trigger: 'add algorithm', 'new trainer', 'new training method', 'implement algorithm'.
Complete workflow for adding a new model adapter. Covers analysis, sample dataclass, adapter implementation (7 abstract methods), registry, example YAML, and verification. Trigger: 'add model', 'support new model', 'integrate model', 'new adapter'.
Complete workflow for adding a new reward model. Covers pointwise vs groupwise design, __call__ contract, registration, YAML config, multi-reward setup, and verification. Trigger: 'add reward', 'new reward model', 'custom reward', 'scoring function'.
Generate images with FLUX models (Black Forest Labs) via inference.sh CLI. Models: FLUX Dev LoRA, FLUX.2 Klein LoRA with custom style adaptation. Capabilities: text-to-image, image-to-image, LoRA fine-tuning, custom styles. Triggers: flux, flux.2, flux dev, flux schnell, flux pro, black forest labs, flux image, flux ai, flux model, flux lora
Generate images with Pruna P-Image models via inference.sh CLI. Models: P-Image, P-Image-LoRA, P-Image-Edit, P-Image-Edit-LoRA. Capabilities: text-to-image, image editing, LoRA styles, multi-image compositing, fast inference. Pruna optimizes models for speed without quality loss. Triggers: pruna, p-image, pruna image, fast image generation, optimized flux, pruna ai, p image, fast ai image, economic image generation, cheap image generation
Generate and edit images with Alibaba Qwen-Image-2.0 models via inference.sh CLI. Models: Qwen-Image-2.0 (fast), Qwen-Image-2.0-Pro (professional text rendering). Capabilities: text-to-image, multi-image editing, complex text rendering. Triggers: qwen image, qwen-image, alibaba image, dashscope image, qwen image 2, qwen image pro
用于 Storyboard Copilot 中“新增 AI 供应商”或“在现有供应商下新增模型”的工程化实现与校验。 当需求涉及 `src/features/canvas/models/**`、`src/features/canvas/models/providers/**`、`src-tauri/src/ai/providers/**`、`build_default_providers()`、`resolveRequest`、`model_aliases`、`inventory::submit!`、设置页 API Key 适配、默认模型切换或模型别名兼容时触发此技能。
Background memory consolidation and learning graduation — overnight knowledge lifecycle.
Security threat model: scan toolkit for attack surface, supply-chain risks.
OmniStudio Data Mapper (formerly DataRaptor) creation and validation with 100-point scoring. Use when building Extract, Transform, Load, or Turbo Extract Data Mappers, mapping Salesforce object fields, or reviewing existing Data Mapper configurations. TRIGGER when: user creates Data Mappers, configures field mappings, works with OmniDataTransform metadata, or asks about DataRaptor/Data Mapper patterns. DO NOT TRIGGER when: building Integration Procedures (use sf-industry-commoncore-integration-procedure), authoring OmniScripts (use sf-industry-commoncore-omniscript), or analyzing cross-component dependencies (use sf-industry-commoncore-omnistudio-analyze).
