Skill UI

This guide outlines the complete MLOps workflow on Databricks. It covers feature engineering using Feature Store, tracking experiments with MLflow, managing model versions in the Model Registry, and deploying real-time inference endpoints via Mosaic AI Model Serving. Ideal for building robust, production-grade ML pipelines.

A comprehensive workflow for implementing robust LLM evaluation using Langfuse. This skill guides users through key processes including scoring traces (numeric, categorical, boolean), collecting user feedback, managing versioned prompts, creating structured datasets, running automated experiments, and implementing LLM-as-a-Judge patterns for advanced quality assurance and system benchmarking.

Automated guidance for managing model checkpoints in ML training workflows, offering best practices, sample code, and validation to keep experiments stable and reproducible.

Automated guidance for model explainability tooling within ML training workflows, giving step-by-step best practices, production-ready code, and validation for experiments whenever users mention explainability tool needs.

Pyvene is a declarative framework designed for performing causal interventions on PyTorch neural networks. It allows researchers to conduct advanced experiments such as activation patching, causal tracing (ROME-style), and interchange intervention training (IIT). Use this library when you need to test causal hypotheses about model behavior, deeply interpret model components, or ensure reproducibility in advanced AI research.

TensorBoard delivers a unified interface to track training metrics, debug network behavior, compare experiments, inspect model graphs, and profile performance for PyTorch or TensorFlow workflows.

Denario orchestrates a multi-agent AI to automate research workflows, covering idea generation, methodology development, computational experiments, and LaTeX publication output.

Stable Baselines3 is a robust, PyTorch-based library providing production-ready implementations of core reinforcement learning algorithms (PPO, SAC, DQN, etc.). It facilitates the training of single-agent RL models, supports the creation of custom Gymnasium environments, and handles advanced parallel training via vectorized environments. It is ideal for rapid prototyping and rigorous RL experiments.

This skill orchestrates end-to-end autonomous AI research projects using a sophisticated two-loop architecture. It manages the entire research lifecycle, from initial literature review and hypothesis formation to running constrained experiments and synthesizing findings into publishable papers. It acts as a project manager, autonomously driving the process while structuring the workspace and presenting progress to human users.

Get a deep, multi-round, and adversarial critical review of your research work, ideas, or drafts. This skill simulates the perspective of a senior ML reviewer (e.g., NeurIPS/ICML level) using dedicated external backends (Codex or Manual). It is designed to find logical gaps, missing experiments, and narrative weaknesses, providing brutal honesty to elevate your academic contribution.

This skill guides users on setting up and running LLM-as-judge evaluations using Arize. It covers defining evaluators (including prompts, classification choices, and models) and executing tasks against real data spans or experiments. It supports key metrics like hallucination, faithfulness, and correctness, and facilitates continuous monitoring for comprehensive LLM performance analysis.

A persistent, graph-based knowledge base designed to accumulate structured knowledge throughout the entire research lifecycle. It stores and connects diverse entities—including published papers, novel ideas, conducted experiments, and testable claims. Inspired by LLM Wiki patterns, it ensures knowledge compounding, meaning the wiki becomes progressively smarter with every piece of input, rather than requiring re-derivation for every query.