Data Science & ML Agent Skills

IGV Integration

Automates IGV snapshot generation for visualizing genomic alignments and variant calls in BAM files.

Gemini Image Generator

Generates high-quality images from text prompts using Google Gemini 3 Pro via the fal.ai API.

Google Gemini Embeddings

Builds production-ready RAG systems and semantic search using optimized Gemini embedding-001 models and vector storage patterns.

DSPy.rb Expert

Builds type-safe, composable LLM applications in Ruby using the DSPy framework to program AI behavior instead of manual prompting.

scverse & Scanpy Essentials

Streamlines single-cell transcriptomics workflows using standardized scverse, AnnData, and Scanpy implementation patterns.

Advanced LLM Evaluation

Implements sophisticated LLM-as-judge methodologies to evaluate and compare AI model outputs with high reliability and bias mitigation.

Advanced LLM Evaluation

Implements production-grade LLM-as-a-judge patterns to evaluate model outputs with high reliability and bias mitigation.

Streamlit for Snowflake

Builds, configures, and deploys native Streamlit data applications directly within the Snowflake Data Cloud.

Memory Systems Design

Implements sophisticated, multi-layered memory architectures including knowledge graphs and temporal persistence for autonomous AI agents.

Memory System Design

Implements advanced memory architectures for AI agents to maintain session continuity and manage structured entity relationships.

Vibe Agent Toolkit (VAT)

Builds and packages portable AI agents that operate across multiple LLM frameworks and deployment targets without vendor lock-in.

Plotly Scientific Visualization

Generates interactive statistical and scientific visualizations using Plotly Express and Graph Objects APIs.

Structural Biology Analysis

Analyzes and validates protein structures, interprets AlphaFold predictions, and performs comparative molecular modeling.

OpenAI Agents SDK Integration

Builds robust AI applications using OpenAI's Agents SDK with multi-agent orchestration, voice capabilities, and advanced error prevention.

Gemini File Search & RAG Integration

Implements Google Gemini File Search to build managed RAG systems with automatic document chunking and semantic search.

ML PRD Generator

Generates structured, evidence-driven Product Requirements Documents (PRDs) specifically tailored for Machine Learning workflows and experiments.

PyMC Bayesian Modeling

Builds and validates sophisticated Bayesian probabilistic models using the PyMC library for advanced statistical inference.

Clinical Value Evaluator

Systematically assesses medical research proposals to quantify their impact on patient outcomes, clinical decision-making, and healthcare systems.

Adaptyv Protein Engineering

Automates protein testing and validation through a cloud laboratory platform for high-throughput protein design workflows.

Open WebUI Management

Manages Open WebUI instances via Podman to provide a browser-based chat interface for Ollama LLM models.

Ollama REST API Integrator

Facilitates direct REST API operations for Ollama using Python to manage models and execute generation tasks.

LocalAI Inference Manager

Manages LocalAI services via Podman to provide OpenAI-compatible local model inference with full GPU acceleration.

PyTDC Therapeutics Data Commons

Provides AI-ready datasets and benchmarks for drug discovery, including ADME, toxicity, and molecular generation tasks.

Biopython Molecular Biology

Streamlines computational molecular biology tasks including sequence analysis, biological file parsing, and genomic database integration.

DrugBank Database Integration

Accesses and analyzes comprehensive pharmaceutical data, including drug properties, interactions, molecular targets, and chemical structures from DrugBank.

Metabolomics Workbench Integration

Accesses the NIH Metabolomics Workbench database to query metabolite structures, experimental study data, and standardized nomenclature.

Ollama Local LLM Management

Manages local LLM inference using Ollama and Podman Quadlet with full GPU acceleration support.

AgentDB Vector Search

Provides ultra-fast semantic vector search and intelligent document retrieval using AgentDB's high-performance HNSW indexing.

FluidSim CFD

Conducts high-performance computational fluid dynamics (CFD) simulations using Python-based pseudospectral methods and MPI parallelization.

Jupyter ML Environment

Manages multi-instance JupyterLab environments with hardware-accelerated GPU support via Podman Quadlet.