3. Data, Science & AI

Extracts text, tables, and metadata from PDF documents for data processing and analysis.

Manages multi-turn dialogues with context retention, Agent coordination, and RAG integration for AI chatbot and customer support development.

Filters incoming questions, rejecting unsuitable queries and routing only decision-focused questions to downstream agents.

Enhances Claude AI model outputs through expert prompt engineering techniques and best practices.

Manages database operations including schema design, query optimization, migrations, and administration for MongoDB and PostgreSQL.

Provides configuration guidance for Wan I2V image-to-video model using ComfyUI, with emphasis on low VRAM optimization and GGUF model deployment.

Processes and generates multimedia content including audio, images, video, and documents using Google Gemini API for multimodal AI tasks.

Optimizes simple prompts to high-quality inputs for Gemini 3 Pro Image generation and AI painting, adhering to Nano Banana Pro best practices.

Creates and manages Obsidian Bases (.base files) with database-like views, filters, and formulas for structured note organization.

Enables development of agentic and multi-agent systems using Google ADK in Python for AI-driven software creation.

Enables knowledge graph-based search of legal documents, precedents, and committee decisions via natural language queries.

Automates data analysis workflows with tools for processing, transforming, and visualizing datasets efficiently.

Analyzes user coding sessions to identify patterns and preferences, then updates the assistant's skill files for continuous improvement.

Simple API for chatting with locally hosted language models via vLLM, enabling message sending and response generation.

Guides users in assessing AI model performance through custom scenarios with structured evaluation criteria.

Evaluates and implements visualization algorithms for ocean/foam rendering, including contour algorithms and performance optimization.

Enhances image resolution using Vertex AI, providing higher-quality outputs for larger displays or print applications.

Generates AI art images via Gemini 3 Pro API, handling prompt rewriting and image creation for user requests.

Performs AI-driven image segmentation to identify, extract, and generate object masks within images using computer vision models.

Applies wave physics principles to simulation code in wave-related source directories, handling wave behavior, shoaling, breaking, and foam physics.

Analyzes images, PDFs, charts, and diagrams to extract and interpret visual information using multimodal AI techniques.

Loads, activates, and optionally installs AI skills from .skill files using slash command interface.

Adds data visualizations using visx library with DaisyUI theming for trends, comparisons, proportions, and funnel data.

Automates deep learning paper replication by parsing PDF content and generating PyTorch implementation code from academic papers.