Enables TimeGPT model fine-tuning on custom datasets with Nixtla SDK. Guides dataset preparation, job submission, status monitoring, model comparison, and accuracy benchmarking. Activates when user needs TimeGPT fine-tuning, custom model training, domain-specific optimization, or zero-shot vs fine-tuned comparison.
Analyze and transform code for TimeGPT-1 to TimeGPT-2 migration. Use when upgrading TimeGPT version. Trigger with 'migrate to TimeGPT-2' or 'upgrade TimeGPT'.
Expert-level AI implementation, deployment, LLM integration, and production AI systems
Deploy and monitor model API endpoints in Domino. Covers creating prediction endpoints, version management, Grafana dashboards for latency/errors/resources, alerting, and GPU inference with NVIDIA Triton. Use when deploying models as APIs, monitoring production endpoints, or debugging endpoint issues.
Forecast convergence patterns in multi-model consensus scenarios.
Model optimization techniques including hyperparameter tuning, architecture search, training optimization, and performance profiling for ML systems.
Master ML experiment tracking - MLflow, W&B, Neptune, versioning, reproducibility
Build production-ready classification and regression models with hyperparameter tuning
Graph Neural Networks (PyG). Node/graph classification, link prediction, GCN, GAT, GraphSAGE, heterogeneous graphs, molecular property prediction, for geometric deep learning.
Real-time monitoring and detection of adversarial attacks and model drift in production
Extract, learn, and integrate PR feedback into the Violet brain
Deep neural network architectures including CNNs, RNNs, Transformers, and modern architectures for vision, NLP, and multimodal tasks.
Execute mechanistic interpretability experiments from JSON specs - family sweeps, itemsets, interactions, minimal cores, validation
Integrate adaptive temporal analysis for drift detection.
Track ML experiments with proper logging and reproducibility. Use when training models or running experiments.
Run credibility checks on feature interpretations including split-half stability and shuffle null tests
Test AI training pipelines for data poisoning vulnerabilities and backdoor injection
Build and train neural networks with PyTorch - MLPs, CNNs, and training best practices
ML framework best practices for PyTorch, TensorFlow, scikit-learn, and modern ML libraries including training patterns and optimization.
Graph Neural Networks (PyG). Node/graph classification, link prediction, GCN, GAT, GraphSAGE, heterogeneous graphs, molecular property prediction, for geometric deep learning.
Implement native Rust ML inference with Candle framework. Use when building GPU-accelerated ML pipelines without Python dependencies.
Supervised & unsupervised learning, scikit-learn, XGBoost, model evaluation, feature engineering for production ML
Expert guidance for PyTorch development covering Deep Reinforcement Learning and NLP Transformers. This skill provides comprehensive knowledge for building RL agents with TorchRL (DQN, PPO) and NLP systems with HuggingFace Transformers. Use this skill when working with PyTorch 2.7+, implementing reinforcement learning algorithms, fine-tuning transformer models, or deploying ML systems to production. Includes current best practices, verified library versions (Dec 2025), and warnings about deprecated APIs.
