Data Science & ML Agent Skills

ReasoningBank Intelligence

Implements adaptive learning and meta-cognitive capabilities to help AI agents optimize strategies through experience and pattern recognition.

Agentica Multi-Agent Spawn

Instantiates sophisticated multi-agent architectures to handle complex reasoning, research, and implementation tasks.

AgentDB Vector Search

Implements high-performance semantic vector search and intelligent document retrieval for RAG-based Claude Code workflows.

AgentDB Memory Patterns

Manages persistent AI agent memory and reasoning patterns using high-performance vector storage and learning algorithms.

Claude RAG Architect

Designs and implements robust Retrieval-Augmented Generation (RAG) architectures using OpenClaw-native execution patterns.

RAG Architecture Engineering

Architects and implements production-grade Retrieval-Augmented Generation systems using specialized engineering guidance and explicit verification.

Prompt Engineering Patterns

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

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.

Portfolio Risk Metrics Calculation

Calculates comprehensive financial risk metrics including VaR, CVaR, Sharpe, and Sortino ratios for quantitative portfolio management.

LLM Training & Fine-Tuning Coach

Guides users through the complete lifecycle of LLM fine-tuning, from initial goal setting to selecting optimal training strategies and models.

ML Pipeline & MLOps Workflow

Orchestrates end-to-end MLOps pipelines from data ingestion and preparation to model training, validation, and production deployment.

Gemini 3 Multimodal Input Processing

Analyzes and extracts insights from images, videos, audio, and PDF documents using Gemini 3 Pro's native multimodal capabilities.

Gemini 3 Multimodal Processing

Processes and analyzes multimodal inputs including images, video, audio, and PDFs using the Gemini 3 Pro model.

AI Embedding Strategies

Optimizes embedding model selection and chunking strategies to enhance semantic search and RAG application performance.

Data Storytelling

Transforms raw analytics into persuasive business narratives through structured storytelling, visualization techniques, and executive-ready frameworks.

Spark Optimization

Optimizes Apache Spark jobs through advanced partitioning, memory management, and shuffle performance tuning.

Bayesian Updating

Refines confidence in hypotheses and technical decisions by systematically weighting prior beliefs against new evidence.

FiftyOne Dataset Inference

Automates the creation of FiftyOne datasets from local media files and executes machine learning model inference pipelines.

LangGraph Fine-Tuning

Optimizes LangGraph application performance by iteratively refining prompts and node-level processing logic based on quantitative evaluation criteria.

LLM Performance Evaluation

Implements comprehensive evaluation strategies for LLM applications using automated metrics, human feedback, and benchmarking.

KBase Datalake Query

Enables seamless SQL querying and exploration of biological data within the KBase/BERDL data lake via MCP.

RAG Systems Engineer

Architects and optimizes high-performance Retrieval-Augmented Generation systems and vector search pipelines.

Trust RL Model Predictions

Optimizes trading strategy execution by removing redundant heuristic pattern filters that conflict with Reinforcement Learning model signals.

Google Veo Video Generation

Generates high-fidelity videos and synchronized audio using Google Veo 3.1 via the Vertex AI API.

Multiplex Imaging Analysis

Optimizes segmentation, feature extraction, and spatial analysis workflows for high-dimensional multiplex immunofluorescence imaging data.

Python Dataclass Optimizer

Optimizes Python dataclasses for memory efficiency, immutability, and validation using advanced PEP 557 patterns.

RL Trading Reward Function Optimizer

Eliminates HOLD bias in reinforcement learning trading models by calibrating reward functions and slippage penalties.

Trading Prediction Outcome Tracker

Automates the recording of actual trading results against predicted signals to enable accurate performance metrics and model retraining.

Multi-Timeframe RL Training

Trains Reinforcement Learning models across multiple market timeframes with automated data resampling and professional market alignment.