3. Data, Science & AI

Detects and reports structural, configuration, logic, and epistemic discrepancies against predefined baselines to ensure consistency in data and model environments.

Provides CLI access to Google's Gemini model for cross-validating responses, leveraging real-time search grounding, and managing multi-turn conversations.

AI-driven portfolio risk management tool for position sizing, risk metric calculation, and dynamic rebalancing strategies.

Sets up, improves, and audits analytics tracking systems including Google Analytics 4 and Google Tag Manager for data collection and measurement.

AI/ML orchestrator for LLM integration, RAG systems, agent-based workflows, and multimodal processing.

Creates optimized llms.txt documentation files to make repositories easily readable by LLMs and AI systems.

Coordinates parallel AI agent swarms to claim, execute, and report on the 'Beads' project, enabling efficient multi-agent task management.

Enables creation, editing, and analysis of spreadsheets with formulas, formatting, and data visualization for data processing tasks.

Guides users on leveraging Sirv AI Studio for AI-powered image and video processing tasks including background removal, upscaling, and generation within e-commerce workflows.

Enforces a coordination model where the main agent delegates all tasks to subagents, ensuring efficient workflow management in AI agent systems.

Provides structured patterns for integrating Google Gemini API with JSON mode and configuration for AI text generation and feature implementation.

Tracks AI token usage and enforces budget limits to prevent cost overruns in LLM projects.

Generates new AI agent skills adhering to the Agent Skills specification, enabling structured development of AI agent capabilities through user-defined instructions and patterns.

Provides a beginner's guide to using the edgartools library for accessing SEC EDGAR financial filings via Python API.

Provides schema design, indexing, and migration guidance for MongoDB databases.

Provides expert guidance on NoSQL database design (Cassandra, DynamoDB), emphasizing query-first modeling and avoiding hot partitions in high-scale systems.

Automates database state switching between clean and populated for testing with automatic backup and restore.

Switches between predefined AI agent profiles (coder, reviewer, planner) to dynamically adjust capabilities and workflows in Claude's agent environment.

Guides construction of stateful, multi-actor AI agents using LangGraph, covering graph structures, node/edge definitions, and state management.

Assists in building AI agents, chatbots, and text generation using an AI SDK, covering API usage, tool calling, and structured output for providers like OpenAI.

Safely copies databases between development, staging, and production environments for testing, recovery, and synchronization.

Teaches the BAML language for defining structured outputs in LLM applications, enabling precise data extraction and formatting.

Analyzes datasets using Python (pandas, numpy) to generate statistical summaries, identify patterns, and create visualizations for CSV/Excel files.

Optimizes token usage and manages context windows for efficient Large Language Model (LLM) interactions and cost reduction.