Data Science & ML Claudeスキル

RLHF Reward Model Training

Trains reward models for RLHF pipelines using preference datasets and sequence classification heads.

Deterministic Receipt Parser Engineer

Develops and maintains high-accuracy receipt and invoice parsers using automated OCR routing and golden fixture testing workflows.

OpenAI Platform Integration

Facilitates the development of AI-powered applications using the OpenAI SDK, covering GPT-5 models, Responses API, and advanced tool calling.

LangGraph Agent Builder

Builds sophisticated, stateful multi-agent workflows and cyclic AI graphs using LangGraph for complex orchestration.

Weaviate Vector Database

Streamlines the integration of Weaviate vector databases to enable semantic search, hybrid queries, and Retrieval-Augmented Generation (RAG) in AI applications.

PyTorch Debugger

Troubleshoots and resolves complex PyTorch issues including CUDA memory errors, tensor shape mismatches, and gradient instabilities.

Observable Plot Visualization

Create expressive, declarative data visualizations and charts using the Observable Plot grammar of graphics directly within your development workflow.

Pandas Debugging & Optimization

Systematically identifies and resolves Pandas DataFrame errors, memory bottlenecks, and data manipulation bugs using the OILER framework.

TensorFlow Debugger

Diagnoses and resolves complex machine learning issues in TensorFlow and Keras workflows systematically.

PyTorch Refactoring & Optimization

Refactors PyTorch code to improve maintainability, performance, and adherence to PyTorch 2.x best practices.

Pandas Refactoring & Optimization

Refactors Pandas code to improve performance, readability, and memory efficiency using modern best practices and 2.0+ features.

Orchestration Patterns

Designs and implements sophisticated multi-agent orchestration architectures for scalable, isolated, and efficient AI workflows.

Advanced RAG Implementation

Builds sophisticated Retrieval-Augmented Generation systems to ground LLMs with external data and private knowledge bases.

Prompt Engineering Patterns

Optimizes LLM performance and reliability through advanced techniques like few-shot learning, chain-of-thought reasoning, and structured template design.

Hugging Face Spaces Manager

Manages and interacts with Hugging Face Spaces by providing tools to search repositories, retrieve metadata, and control application runtimes.

Prompt Engineering Patterns

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

Ensembl Genome Database Integration

Simplifies access to the Ensembl REST API for comprehensive genomic data retrieval, sequence analysis, and variant prediction across 250+ species.

Medchem Tools

Applies medicinal chemistry filters and drug-likeness rules to prioritize compound libraries and assess molecular quality.

Statsmodels Statistical Modeling

Provides a comprehensive suite of tools for statistical estimation, inference, and rigorous econometric analysis in Python.

AnnData Genomics Data Management

Manages annotated data matrices and single-cell genomics workflows using the AnnData framework and scverse ecosystem.

R Data Science & Shiny Development

Accelerates R 4.4+ development with expert patterns for the tidyverse, ggplot2 visualizations, and Shiny web applications.

USPTO Database & IP Analysis

Accesses USPTO APIs to perform comprehensive patent and trademark searches, retrieve examination histories, and analyze intellectual property data.

Ollama Client

Interfaces with the Ollama API to perform high-performance text completions and clinical analysis using the Phi-4 language model.

PDB Database

Accesses and manages 3D biological macromolecular structures from the RCSB Protein Data Bank for structural biology and drug discovery.

Swarm Orchestration

Orchestrates multi-agent AI swarms for parallel task execution and dynamic workflow management using agentic-flow.

ReasoningBank Intelligence

Implements adaptive learning systems to enable AI agents to recognize patterns, optimize strategies, and improve through experience.

AgentDB Reinforcement Learning

Enables autonomous agents to learn and improve through experience using nine advanced reinforcement learning algorithms.

Grail Miner Management

Optimizes the deployment, operation, and performance of Grail miners on Bittensor Subnet 81 for verifiable language model post-training.

Experiment Tracking

Streamlines machine learning experiment management by integrating tracking tools like MLflow and W&B into development workflows.

NeMo Evaluator SDK

Evaluates Large Language Models across 100+ benchmarks using a reproducible, containerized framework.