Data Science & ML Agent Skills

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.

Nixtla Exogenous Integrator

Integrates external variables like holidays and weather data into TimeGPT models to significantly improve time series forecasting accuracy.

Biomni Biomedical AI Agent

Automates complex biomedical research tasks including genomics analysis, drug discovery, and CRISPR screening using integrated data and code execution.

Ollama Local LLM Management

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

ML Pipeline & MLOps Workflow

Orchestrates end-to-end MLOps pipelines from data preparation and model training to production deployment and monitoring.

DrugBank Database Analyst

Accesses and analyzes comprehensive pharmaceutical data from DrugBank for drug discovery, interaction analysis, and pharmacological research.

Jupyter ML Environment

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

Ollama OpenAI Bridge

Facilitates the use of local Ollama models with the official OpenAI Python library and compatible AI orchestration frameworks.

QIG Purity Guardian

Enforces E8 architecture standards and QIG purity protocols within the Pantheon development ecosystem.

Biomni Biomedical Research Agent

Executes autonomous biomedical research tasks across genomics, drug discovery, and molecular biology using integrated databases and code execution.

RAG Implementation Guide

Build robust Retrieval-Augmented Generation (RAG) systems using vector databases and semantic search to ground AI responses in external data.

FiftyOne Dataset Manager

Manages FiftyOne dataset visualization and curation environments using Podman Quadlet containers with integrated MongoDB sidecars.

ComfyUI Stable Diffusion Manager

Manages ComfyUI instances for node-based Stable Diffusion image generation with automated GPU configuration and model management.

ML Pipeline Workflow

Builds end-to-end MLOps pipelines covering data preparation, model training, validation, and production deployment.

Vector Index Tuning

Optimizes vector database performance by balancing search latency, recall accuracy, and memory footprint.

Rust Machine Learning & AI

Guides the development of high-performance ML and AI applications in Rust using memory-efficient patterns and GPU acceleration.

QLoRA Experimentation & Optimization

Optimizes LLM fine-tuning via advanced QLoRA patterns, hyperparameter tuning, and memory-efficient implementation strategies.

Book SFT & Style Transfer Pipeline

Builds high-quality fine-tuning datasets from literary works to train AI models in specific authorial voices and writing styles.

Context Degradation Diagnostics

Diagnoses and mitigates AI agent performance failures caused by long-context attention loss, poisoning, and informational clash.

Multi-Agent Architecture Patterns

Designs and implements sophisticated multi-agent systems using supervisor, swarm, and hierarchical patterns to solve complex context management challenges.

Ollama Python Integration

Provides a clean, Pythonic interface for local LLM inference, chat completions, and model management using the official Ollama library.

Context Engineering Fundamentals

Provides foundational expertise in context engineering to optimize AI agent performance and manage token usage effectively.

GRPO Alignment & RLHF

Implements Group Relative Policy Optimization for efficient LLM alignment and reinforcement learning from human feedback.

PEFT & LoRA Fine-Tuning

Streamlines parameter-efficient fine-tuning for LLMs using LoRA, QLoRA, and Unsloth to optimize memory and training speed.

DPO Preference Optimization

Optimizes Large Language Models using Direct Preference Optimization to align behavior with preferred response pairs without explicit reward modeling.

Vision Model Fine-Tuning

Fine-tunes vision-language models like Pixtral and Ministral using Unsloth's FastVisionModel optimizations for faster training.

BDI Mental State Modeling

Transforms external RDF context into formal Belief-Desire-Intention (BDI) models to enable rational agency and explainable reasoning in AI agents.

Model Quantization

Optimizes large language models for efficient inference and training by reducing memory footprint using advanced precision-shifting techniques like 4-bit and 8-bit quantization.