llm-dev

What problem does it solve?

This Skill helps you design, evaluate, and ship large language model features without guesswork, covering the full path from model selection to production readiness.

Core Features & Use Cases

• LLM Architecture Guidance: Understand transformer internals, attention variants, positional encoding, tokenization, and mixture-of-experts tradeoffs.
• Training and Alignment: Choose and apply pretraining, fine-tuning, and preference optimization methods such as LoRA, QLoRA, DPO, and RLHF.
• RAG, Agents, and Tool Use: Build retrieval-augmented systems, structured tool workflows, MCP integrations, and agentic patterns with clear failure boundaries.
• Evaluation and Production: Define eval-driven development plans, measure quality, control latency and cost, and harden systems for observability and safety.

Quick Start

Ask for an end-to-end plan for an LLM feature, including architecture, evaluation strategy, implementation boundaries, and production risks.