Home / Stacks / LLM Ops & Training
🧠

LLM Ops & Training

Take models from training to production — fine-tune with LoRA/QLoRA, serve at high throughput with vLLM, track experiments reproducibly, and monitor cost and quality in production with Langfuse

5 skills · Works with Claude Code, Codex, Cursor & more

⚙️ Engineering
RARE

LLM Fine-Tuning Expert

Fine-tune LLMs with LoRA, QLoRA, and PEFT. Prepares JSONL datasets, sets hyperparameters, runs adapter training and instruction tuning, and applies RLHF/DPO — then quantizes and deploys the resulting model.

Community 3.2K
Scanned
fine-tuning lora qlora
mkdir -p ~/.claude/skills/fine-tuning-expert && curl -fsSL https://raw.githubusercontent.com/Jeffallan/claude-skills/HEAD/skills/fine-tuning-expert/SKILL.md -o ~/.claude/skills/fine-tuning-expert/SKILL.md
⚙️ Engineering
RARE

vLLM Model Serving

Serve LLMs at high throughput with vLLM. Configures PagedAttention, continuous batching, tensor parallelism, and OpenAI-compatible endpoints, plus GPTQ/AWQ/FP8 quantization for serving on limited GPU memory.

Community 2.6K
Scanned
vllm inference serving
mkdir -p ~/.claude/skills/vllm-serving && curl -fsSL https://raw.githubusercontent.com/diegosouzapw/awesome-omni-skill/HEAD/skills/data-ai/serving-llms-vllm/SKILL.md -o ~/.claude/skills/vllm-serving/SKILL.md
⚙️ Engineering
RARE

ML Experiment Tracking

Track ML experiments with reproducible parameters, metrics, and artifacts. Logs runs, compares hyperparameter sweeps, and versions models so training results stay reproducible and easy to compare.

Community 1.5K
Scanned
mlops experiment-tracking reproducibility
mkdir -p ~/.claude/skills/ml-experiment-tracking && curl -fsSL https://raw.githubusercontent.com/diegosouzapw/awesome-omni-skill/HEAD/skills/machine-learning/ml-experiment-tracking/SKILL.md -o ~/.claude/skills/ml-experiment-tracking/SKILL.md
⚙️ Engineering
RARE

scikit-learn ML

Build classical ML pipelines with scikit-learn. Covers classification, regression, clustering, preprocessing, feature selection, model selection, and cross-validation with a consistent fit/transform/predict API.

Community 3.8K
Scanned
scikit-learn machine-learning python
mkdir -p ~/.claude/skills/scikit-learn && curl -fsSL https://raw.githubusercontent.com/diegosouzapw/awesome-omni-skill/HEAD/skills/machine-learning/scikit-learn/SKILL.md -o ~/.claude/skills/scikit-learn/SKILL.md
⚙️ Engineering
RARE

Langfuse LLM Observability

Instrument LLM apps with Langfuse for tracing, prompt management, evaluation, and datasets. Captures nested traces across LangChain, LlamaIndex, and OpenAI calls to debug latency, cost, and quality in production.

Community 2.4K
Scanned
langfuse observability tracing
mkdir -p ~/.claude/skills/langfuse && curl -fsSL https://raw.githubusercontent.com/diegosouzapw/awesome-omni-skill/HEAD/skills/ai-agents/langfuse/SKILL.md -o ~/.claude/skills/langfuse/SKILL.md

More Stacks

🤝

Agent Engineering Kit

Engineer reliable AI agents — build on the Claude Agent SDK, coordinate teams with multi-agent patterns, give them memory, and keep them safe with mechanical guardrails and evaluation
🕹️

AI Agent Frameworks

Pick the right framework and ship agents fast — type-safe Python with Pydantic AI, multi-agent orchestration with OpenAI Agents SDK, Agno, and CrewAI, TypeScript graphs with LangGraph.js, and typed workflows with Mastra
🤖

AI Builder Stack

Build AI-powered applications with RAG, multi-agent systems, prompt engineering, and MCP integrations

View all stacks →

Added to wishlist