3. Data, Science & AI

Scans song lyrics for explicit content and verifies explicit flags match actual text, ensuring accurate content labeling.

Analyzes student enrollment data to provide insights on counts, course loads, and time-based enrollment trends.

Analyzes student financial data including tuition payments, scholarships, and aid to provide revenue insights, balance tracking, and coverage rate reporting.

Integrates scientific ontologies for term resolution, semantic typing, and federated knowledge systems within the Sounio framework.

Guides engineering leads through AI-SDLC onboarding with readiness analysis, lint rehearsal, and three-month adoption planning for stakeholder artifacts.

Analyzes VCD simulation files to identify signal anomalies, detect clock behavior, and trace signal timing for hardware debugging.

Enables knowledge distillation for efficient AI model training, compressing large models into smaller, faster versions while preserving accuracy.

Delegates tasks to Gemini AI for codebase investigation, web search, and log analysis using iterative CLI or direct MCP modes.

Provides a unified dashboard for real-time business metrics including traffic, revenue, and errors across all products, enabling quick weekly traction checks.

Reduces token count in instruction files (prompts, skills) while preserving full capability for LLM applications.

Orchestrates multiple specialized AI tools to collect, synthesize, and reach consensus on complex analysis, review, audit, and investigation tasks.

Manages context preparation and token budgeting for subagents in multi-agent LLM systems, ensuring efficient information flow and cost control.

Provides confidence assessment, citation verification, and recovery protocols to detect and mitigate hallucinations in AI-generated content, serving as a quality gate.

Analyzes AI skill usage patterns and performance metrics from session transcripts to optimize effectiveness and identify improvement areas.

Provides Gemini AI as an oracle for code review and planning, offering second opinions without implementation tasks.

Refines and appends optimized prompts for large language models to PROMPT.md documentation.

Creates AI tools using Progressive Disclosure, enabling design of rules, skills, agents, and structured workflows.

Fetches YouTube video transcripts using the youtube-transcript-api library for data extraction and analysis.

Preview CSV files in browser with interactive sorting, filtering, and column statistics for efficient data exploration.

Generates interactive 2D data visualizations with D3.js, supporting zoom, pan, and custom rendering for data exploration.

Tracks and graphs token usage across API keys and subscriptions, providing real-time visualizations for cost monitoring and usage analysis.

Generates comprehensive R-based machine learning and AI tutorials covering theory, practice, and key algorithms like random forest and neural networks.

Creates comprehensive R data visualization tutorials with ggplot2, covering chart types, styling, and publication-ready plots through theory and practice.

Installs, configures, and debugs vLLM inference server on NVIDIA GPUs, including tensor parallelism setup and MoE model support.