Explorar skills

Comprehensive growth hacking strategy including growth loops, AARRR pirate metrics, channel prioritization (Bullseye), viral mechanics (K-factor), ICE experiment scoring, and 90-day experimentation roadmap using Growth Loops, Pirate Metrics, and Traction Bullseye frameworks.

Debug common ML training issues (NaN loss, OOM, slow training, convergence problems) and provide solutions. Use when training fails, metrics don't improve, or encountering errors like NaN loss, CUDA OOM, or slow convergence.

Comprehensive guide for PyTorch Lightning - LightningModule, Trainer, distributed training, PyTorch 2.0 torch.compile integration, Lightning Fabric, and production best practices

Run comprehensive code quality checks with ruff (format, lint) and ty (type checking). Use when checking code quality, fixing linting errors, or ensuring code follows best practices before commits or PRs.

Execute training runs with proper monitoring, checkpointing, and experiment tracking. Use when starting training, resuming training, debugging training issues, or setting up multi-GPU/distributed training with PyTorch Lightning and Hydra.

Hugging Face Transformers with PyTorch Lightning - LightningModule integration, distributed training (FSDP/DeepSpeed), PEFT (LoRA/QLoRA), data pipelines with HF Datasets, evaluation metrics, and common NLP tasks

Comprehensive guide for Pixi package manager - Python environment management, CUDA/GPU support, PyPI integration, Docker/Pixi-Pack deployment, and best practices for ML research

Use when implementing features, refactors, or multi-file changes. Orchestrated plan-then-execute with subagent dispatch and regression checkpoints.

Паттерн Schema-Guided Reasoning (SGR) для решения задач на базе LLM с использованием structured output, constrained decoding, Pydantic моделей. Используй когда нужно: структурированный вывод, JSON schema, воспроизводимость рассуждений, тестируемость промежуточных шагов, направленное рассуждение модели, response_format, parse().

Enforces Karpathy's 4 LLM-coding principles (Think Before Coding, Simplicity First, Surgical Changes, Goal-Driven Execution). Auto-fires whenever the user requests writing, editing, refactoring, fixing, or adding code. Blocks drive-by refactoring, over-engineering, and speculative changes; demands clarification before implementation, minimum code, surgical edits, and conversion of tasks into verif

Full-stack implementation agent that writes complete, production-ready code following an approved architecture and schema. Triggers on: write the code, implement features, build the app, code the MVP, generate codebase.

Application security agent that audits code for OWASP Top 10 vulnerabilities, hardcoded secrets, and common security flaws. Triggers on: security audit, vulnerability scan, OWASP check, security review, penetration test, hardening.