Data Science & ML Agent Skills

AgentDB Memory Patterns

Implements high-performance persistent memory and pattern learning for stateful AI agents using AgentDB.

Transfer Learning Adapter

Automates the adaptation and fine-tuning of pre-trained machine learning models for specific tasks and new datasets.

AgentDB Reinforcement Learning Plugins

Deploys and trains autonomous agents using nine reinforcement learning algorithms integrated with the AgentDB ecosystem.

ClinVar Database Integration

Queries the NCBI ClinVar database to identify genetic variants, interpret pathogenicity classifications, and annotate genomic data for bioinformatics pipelines.

AutoML Pipeline Builder

Automates the creation of end-to-end machine learning pipelines including data validation, model selection, and hyperparameter tuning.

Neural Network Builder

Automates the design, configuration, and implementation of complex neural network architectures through natural language commands.

UniProt Database Access

Provides direct REST API access to the UniProt database for protein search, sequence retrieval, and cross-database ID mapping.

BRENDA Enzyme Database

Accesses the comprehensive BRENDA enzyme database to retrieve kinetic parameters, reaction equations, and biochemical data via SOAP API.

AgentDB Vector Search

Implements high-performance semantic vector search and HNSW indexing for intelligent document retrieval and RAG systems.

Dimensions Academic Search

Queries the Dimensions database to find and analyze academic publications, grants, patents, and clinical trials for evidence-based research.

Scientific Brainstorming

Facilitates creative research ideation by generating hypotheses, identifying research gaps, and exploring interdisciplinary connections.

Scholarly Literature Review

Conducts systematic, multi-database academic literature reviews with verified citations and professional formatting.

Scientific Visualization

Generates publication-quality, journal-ready figures and multi-panel scientific visualizations following academic standards.

Scientific Brainstorming

Facilitates creative research ideation and exploratory scientific problem-solving through interdisciplinary analogies and assumption-challenging frameworks.

Data Engineering & Pipeline Optimization

Implements reliable, reproducible data pipelines and infrastructure using best practices for ETL/ELT workflows.

Account-Aware RL Training

Integrates real-time account performance metrics and drawdown penalties into Reinforcement Learning models to optimize trading risk management.

QuTiP Quantum Simulation

Simulates and analyzes open and closed quantum systems using the Quantum Toolbox in Python.

GLM-4.7 Prompting Guide

Optimizes Claude Code interactions with the GLM-4.7 model through advanced prompting patterns and multi-agent ensemble strategies.

FiftyOne ML Dataset Curator

Manages FiftyOne dataset visualization and curation workflows using Podman Quadlet containers with MongoDB persistence.

QLoRA Experimentation & Optimization

Optimizes Large Language Model fine-tuning through advanced QLoRA experimentation patterns, rank selection, and adapter management.

Ollama OpenAI Compatibility

Enables seamless integration between Ollama and the OpenAI ecosystem using a compatible API layer for popular libraries.

Fast AI Model Inference

Accelerates LLM inference by up to 2x using Unsloth and vLLM backends for optimized model execution.

LocalAI Manager

Manages LocalAI inference services with GPU acceleration using Podman Quadlet containers for local model hosting.

HuggingFace Model Importer

Imports GGUF models directly from HuggingFace into Ollama for immediate local inference and testing.

RAG Implementation Guide

Implements Retrieval-Augmented Generation workflows to ground AI responses with external knowledge and private documents.

Open WebUI Manager

Manages and configures Open WebUI instances for interacting with Ollama models via Podman containers.

Jupyter ML/AI Environment

Manages multiple JupyterLab development environments with GPU acceleration using Podman Quadlet containers.

Model Fine-Tuning & Training

Provides expert guidance and implementation patterns for fine-tuning Large Language Models using PyTorch and HuggingFace.

GRPO RLHF Alignment

Optimizes LLM alignment through Group Relative Policy Optimization (GRPO) for stable reinforcement learning and reasoning model training.

Vision Model Fine-Tuning

Fine-tunes and optimizes vision-language models like Pixtral and Ministral using Unsloth's FastVisionModel and LoRA.