3. Data, Science & AI

Provides structured scoring criteria for evaluating AI-generated content quality across naturalness, alignment, and richness dimensions.

Analyzes meeting transcripts and recordings to identify communication patterns, behavioral insights, and actionable feedback for leadership development.

Analyzes Twitter/X backup data with keyword search, popularity tracking, and AI-powered discussion for social media insights.

Enables voice input (STT) and voice output (TTS) for natural interaction with Claude Code, supporting multilingual commands and voice mode activation.

AI-powered tool for visual content analysis, web search, and GitHub code exploration using GLM-4.6V model.

Assembles dialogue components (requests, traces, captions) into a structured JSON format for stage 3 processing in dialogue systems.

Generates structured reasoning traces capturing user intent, inferred details, and step-by-step plans for AI decision transparency.

Extracts structured property data from real estate listings (Zillow, Redfin, etc.) using AI-powered web scraping for downstream processing.

Command-line tool for filtering, transforming, and manipulating JSON data efficiently in data pipelines and scripts.

Provides understanding of data formats, required keys, and metadata structures for Supervised Fine-Tuning (SFT) pipelines in AI model training.

Specializes in BIM-GIS integration and AEC data engineering using Industry Foundation Classes (IFC) schema for digital twin implementations.

Handles MongoDB database operations: connection, queries, indexes, aggregation, transactions, and bulk operations.

Automates Galaxy workflows using BioBlend and Planemo, handling API interactions, status checks, error handling, and batch processing for bioinformatics data analysis.

Optimizes token usage for Claude Code, providing efficient strategies for file handling, command execution, and model selection to reduce costs.

Provides command utilities for Redis data structures including key operations, hashes, pub/sub, streams, and clustering.

Provides reusable utilities for serverless PostgreSQL database operations, connection pooling, and asynchronous connections with Neon MCP server.

Expert in applying statistical models and data analysis to develop trading strategies, manage financial risk, and optimize investment portfolios.

Fetches and formats GitHub issues as markdown for LLM consumption, enabling efficient issue-based code implementation and analysis.

Expert assistance in machine learning engineering, including feature engineering, model debugging, and building production-ready ML pipelines.

Specializes in prediction market operations including market making, odds calculation, and event resolution through data analysis.

Enables database-persisted conversation state management for AI agents, supporting history retention and session resumption after server restarts.

Enables implementation of OpenAI Agents SDK for building intelligent AI assistants with function tools, MCP server integration, and context management.

Guides creation, modification, and registration of AI agents within the VIKI platform for streamlined agent management.

Configures automatic text continuation parameters for AI inference models.