Data Science & ML Agent Skills

ToolUniverse Scientific Research Skill

Empowers AI agents to conduct scientific research by providing standardized access to over 600 bioinformatics, cheminformatics, and genomics tools.

AnnData Scientific Data Management

Manages annotated data matrices for single-cell genomics and large-scale biological datasets using the AnnData Python framework.

AI Evaluation Infrastructure Validator

Validates evaluation contract readiness and metadata integrity before initiating agentic optimization loops.

NCBI Gene Database Explorer

Queries and retrieves comprehensive gene information from NCBI databases for genomic research and functional analysis.

Prompt Engineering Patterns

Implements advanced prompt engineering techniques to maximize LLM performance, reliability, and controllability in production applications.

TorchDrug Molecular Science

Accelerates drug discovery and molecular research using graph neural networks and PyTorch-based machine learning.

ML Pipeline Workflow

Builds and orchestrates end-to-end MLOps pipelines from data preparation through production deployment.

GEO Database Explorer

Accesses and analyzes functional genomics data from the NCBI Gene Expression Omnibus (GEO) repository.

Zarr Python Scientific Data

Manage large-scale N-dimensional arrays with chunking, compression, and cloud-native storage for scientific computing workflows.

Scikit-learn Machine Learning

Implements professional machine learning workflows in Python using scikit-learn for classification, regression, clustering, and data preprocessing.

RAG System Implementation

Builds robust Retrieval-Augmented Generation (RAG) systems using vector databases and semantic search to ground LLM responses in proprietary data.

Arboreto Gene Regulatory Network Inference

Infers gene regulatory networks from transcriptomics data using scalable algorithms like GRNBoost2 and GENIE3.

Biopython Bioinformatics Toolkit

Automates computational molecular biology tasks including sequence manipulation, NCBI database queries, and structural analysis.

PyMC Bayesian Modeling

Build, fit, and validate complex Bayesian probabilistic models using the PyMC Python library and modern MCMC sampling techniques.

LangChain Architecture Design

Architects sophisticated LLM applications using the LangChain framework for agents, memory management, and complex tool integration.

Hypogenic Scientific Discovery

Automates the generation and testing of scientific hypotheses by synthesizing empirical data and existing research literature.

Data Versioning Refactor

Implements lightweight dataset tracking and reproducibility patterns to ensure data changes are explicit and traceable.

PufferLib RL Performance

Accelerates reinforcement learning workflows with high-performance training, optimized environment vectorization, and seamless multi-agent support.

FiftyOne Image Deduplication

Identifies and removes duplicate or visually similar images in FiftyOne datasets using deep learning embeddings.

ReasoningBank with AgentDB

Implements high-performance adaptive learning and experience replay for AI agents using the AgentDB vector engine.

Antigravity Context Manager

Orchestrates dynamic AI context, intelligent memory systems, and RAG workflows for enterprise-scale multi-agent applications.

Astropy Scientific Computing

Performs advanced astronomical data analysis, coordinate transformations, and cosmological calculations using the core Astropy Python library.

Prompt Engineering Patterns

Optimizes LLM performance and reliability through advanced prompting techniques like few-shot learning and chain-of-thought reasoning.

Advanced Excel & Spreadsheet Manager

Manages complex Excel workbooks with automated formula recalculation, professional financial modeling standards, and deep data analysis capabilities.

Prompt Engineering Patterns

Master advanced prompt engineering techniques to maximize LLM performance, reliability, and controllability in production applications.

PyTorch Geometric (PyG) Expert

Builds and trains sophisticated Graph Neural Networks (GNNs) using the PyTorch Geometric library for irregular data structures.

Prompt Engineering Patterns

Implements advanced prompt engineering techniques to optimize LLM performance, reliability, and structured output in production environments.

Nixtla Research Assistant

Monitors and summarizes the Nixtla forecasting ecosystem to provide actionable updates on TimeGPT, StatsForecast, and MLForecast.

Hybrid Search Implementation

Combines vector similarity and keyword-based search to improve retrieval accuracy in RAG systems and search engines.

Vector Similarity Search Patterns

Implement high-performance similarity search and vector retrieval patterns across multiple database providers.