Machine learning

Use when designing how one canonical agent core runs across Codex, Claude Code, Gemini CLI, Cursor, and AGENTS.md-compatible tools.

Use when designing how one canonical agent core runs across Codex, Claude Code, Antigravity, Cursor, and AGENTS.md-compatible tools.

Writes rigorous mathematical proofs for ML/AI theory.

Persistent research knowledge base that accumulates papers, ideas, experiments, claims, and their relationships across the entire research lifecycle.

Run GPU workloads on Modal — training, fine-tuning, inference, batch processing.

Diagnose, troubleshoot, and advise on any Qdrant deployment by loading the latest official Qdrant skills live from skills.qdrant.tech.

科研分析能力入口：统计、因果推断、数据清洗与可视化。Use for statistics, econometrics, wrangling, causal inference, charts, and publication figures.

This skill should be used for project-level decisions about LLM-powered systems: whether an LLM is the right primitive for the task at hand, the shape of a…

Use when building or maintaining a self-contained personal knowledge base (an LLM wiki) in plain markdown.

Generates publication-quality figures for ML papers from research context.

Open-source embedding database for AI applications.

Facebook's library for efficient similarity search and clustering of dense vectors.

Post-training 4-bit quantization for LLMs with minimal accuracy loss.

Compress large language models using knowledge distillation from teacher to student models.

Framework for building LLM-powered applications with agents, chains, and RAG.

Battle-tested PyTorch training recipes for all domains — LLMs, vision, diffusion, medical imaging, protein/drug discovery, spatial omics, genomics.

Reduce LLM size and accelerate inference using pruning techniques like Wanda and SparseGPT.

GPU-accelerated data curation for LLM training.

Managed vector database for production AI applications.

Generates conference presentation slides (Beamer LaTeX PDF and editable PPTX) from a compiled paper with speaker notes and talk script.

Adds PyTorch FSDP2 (fully_shard) to training scripts with correct init, sharding, mixed precision/offload config, and distributed checkpointing.

High-level PyTorch framework with Trainer class, automatic distributed training (DDP/FSDP/DeepSpeed), callbacks system, and minimal boilerplate.

Scalable data processing for ML workloads.

Framework for state-of-the-art sentence, text, and image embeddings.