3. Data, Science & AI

Optimizes AI prompts to enhance clarity, reduce token usage, and improve model output quality through structural refinement.

Extracts structured text from PDFs using AI (Docling) and traditional tools (PyMuPDF, pdfplumber) for LLM and RAG systems.

Provides Anthropic's official best practices for optimizing prompts, debugging outputs, and improving response quality in AI interactions.

Enables CLI-based testing and automation using Codex AI model with reasoning, web search, and multi-turn conversation features.

Generates Wikipedia-style articles with citations using Stanford STORM AI, powered by LLM and search APIs for comprehensive content.

Converts web URLs to LLM-friendly Markdown and performs AI-enhanced web searches using Jina AI APIs for data preparation and retrieval.

Integrates Tavily API for real-time web search results to enhance RAG systems, requiring an API key.

Generates comprehensive research reports through iterative web searches using LangChain and LLMs, requiring API keys for LLM and search services.

Enables real-time web and X (Twitter) search and AI-powered information synthesis using the xAI Grok API, requiring XAI_API_KEY.

Provides real-time, citation-backed answers using Perplexity Sonar API for up-to-date information synthesis and quick research.

Enables code-based agentic research using Hugging Face Smolagents, supporting multiple LLM providers and web search for data gathering.

Provides professional market data access for all asset classes through the DataBento API.

Analyzes and structures prompts using research-backed frameworks to optimize AI model responses, providing expert guidance for effective prompt engineering.

Extracts form field data from PDF documents to enable automated form filling workflows.

Comprehensive LLM prompt optimization workflow, from basic improvements to multi-agent orchestration, using advanced techniques and expert analysis.

Comprehensive system for AI context architecture, memory management, and knowledge engineering with intelligent workflow orchestration.

Automates specialized AI skill creation and optimization using pattern recognition, intelligent tool selection, and automated quality assurance.

Injects structured knowledge into LLMs using semantic frameworks to improve factual accuracy and reasoning capabilities.

Manages storage and retrieval of AI embeddings and memory configurations via API or repository helpers for AI applications.

Optimizes AI prompts for system, agent, and documentation (CLAUDE.md, AGENTS.md, SKILL.md) to enhance LLM effectiveness and agent performance.

Automates PostgreSQL database and user creation on any host, enabling rapid database setup for applications.

Assists in running DNN model checkpoints by locating dependencies, verifying requirements, and generating step-by-step tutorials for reproducible experiments.

Conducts data analysis and generates visualizations using Python for data processing and insights.

Extracts text and tables from PDF documents at scale, enabling efficient analysis by large language models (LLMs).