Explorar skills

Orchestrator for the 5 Peec AI growth skills (peec-setup, peec-content-intel, peec-cluster, peec-outreach, peec-report). Reads project state and last learnings, detects the single highest-leverage gap, and returns exactly one next-move decision with a measurable 4-week metric — then hands off to the skill that executes it. Use when multiple priorities compete and you need one call, not a dashboard

Runs LLM inference on CPU, Apple Silicon, and consumer GPUs without NVIDIA hardware. Use for edge deployment, M1/M2/M3 Macs, AMD/Intel GPUs, or when CUDA is unavailable. Supports GGUF quantization (1.5-8 bit) for reduced memory and 4-10× speedup vs PyTorch on CPU.

Fine-tune LLMs using reinforcement learning with TRL - SFT for instruction tuning, DPO for preference alignment, PPO/GRPO for reward optimization, and reward model training. Use when need RLHF, align model with preferences, or train from human feedback. Works with HuggingFace Transformers.

Serves LLMs with high throughput using vLLM's PagedAttention and continuous batching. Use when deploying production LLM APIs, optimizing inference latency/throughput, or serving models with limited GPU memory. Supports OpenAI-compatible endpoints, quantization (GPTQ/AWQ/FP8), and tensor parallelism.

Parameter-efficient fine-tuning for LLMs using LoRA, QLoRA, and 25+ methods. Use when fine-tuning large models (7B-70B) with limited GPU memory, when you need to train <1% of parameters with minimal accuracy loss, or for multi-adapter serving. HuggingFace's official library integrated with transformers ecosystem.

Search the Obsidian Zettelkasten vault using qmd hybrid search (BM25 + vector + LLM re-ranking). Use when the user wants to find notes, quotes, people, books, literature notes, permanent notes, or any content in their personal knowledge base.

Complete AI-native product management — AI feature strategy, model selection, evaluation frameworks, AI UX design, responsible AI, and building products that use LLMs, CV, and ML as core features

Grimoire converte PDFs, documentos ou texto bruto em notas de leitura e um pacote de habilidades reutilizável em uma única análise. Ele classifica automaticamente o conteúdo, estrutura notas específicas para o Obsidian e extrai habilidades, utilizando a API MinerU sem LLMs ou escrita direta no vault.

Weekly / monthly closed-loop reporter for Peec AI visibility growth. Measures what moved (visibility per prompt, cluster, zone) against what was invested (content published, pitches sent, forum answers), detects winning patterns, and outputs a ranked next-actions list — not a dashboard. Closes the feedback loop for the growth agent. Use weekly for active projects or monthly for maintenance-mode.

Path to the LLM Wiki knowledge base directory

Search personal knowledge bases, notes, docs, and meeting transcripts locally using qmd — a hybrid retrieval engine with BM25, vector search, and LLM reranking. Supports CLI and MCP integration.

Simple Preference Optimization for LLM alignment. Reference-free alternative to DPO with better performance (+6.4 points on AlpacaEval 2.0). No reference model needed, more efficient than DPO. Use for preference alignment when want simpler, faster training than DPO/PPO.