3. Data, Science & AI

Comprehensive review of scientific documents evaluating claims, evidence, methodology, and presentation to ensure research quality and clarity.

Comprehensive toolkit for computational materials science, enabling crystal structure analysis, Materials Project API integration, and DFT input generation.

Enables structured scientific documentation in org-mode with date-based organization for hypotheses, methods, and reproducible research outcomes.

Generates and edits images from text using AI models including Google Gemini, Imagen, and OpenAI for visual asset creation.

Expert assistant for calculating materials properties using first-principles methods with ASE, including structure relaxation and thermodynamic modeling.

Provides expert guidance for JAX, a high-performance numerical computing library with automatic differentiation, JIT compilation, and GPU/TPU acceleration for AI/ML development.

Provides diagnostic workflow and reporting for context health checks in data environments, ensuring consistency, reducing duplicates, and optimizing data quality.

Comprehensive tool for creating, editing, and analyzing spreadsheets with formula support, formatting, and data visualization capabilities.

Guides safe selection and application of reasoning controls for AI models, including depth adjustments and metric visibility.

Provides database schema design patterns and optimization strategies for relational and NoSQL databases to enhance data persistence and query performance.

Consults multiple external LLMs (Gemini, OpenAI, Qwen) to provide second opinions, alternative solutions, and delegated task execution.

Manages local LLM operations using Ollama on Apple Silicon, including model deployment, chat interfaces, and performance diagnostics.

Provides vector search indexing and querying workflows with MCP Vector Search, including setup, reindexing, and auto-index strategies.

Automates SEC EDGAR filing extraction, CIK-based discovery, and report generation through a recipe-driven pipeline.

Framework for building LLM-powered applications with chains, agents, RAG, and memory management capabilities.

Manages multi-turn conversations efficiently using AI summarization and embedding-based context retrieval.

Framework for building stateful, multi-agent AI applications with cyclical workflows, human-in-the-loop patterns, and persistent checkpointing.

Enables programmatic reading, writing, and manipulation of Excel spreadsheets for data extraction and processing tasks.

Platform for building, training, and deploying AI agents on DigitalOcean's GPU infrastructure with foundation models and knowledge bases.

Generates timestamped reports in CSV, JSON, and Markdown formats with built-in summaries and post-processing capabilities for data exports.

Unified API for accessing 200+ LLMs with intelligent routing, cost optimization, and streaming capabilities.

Automates AI prompt engineering by treating prompts as code, with systematic evaluation and compilers for optimization.

Manages project-specific knowledge for AI prompt optimization, including schema, extraction, and retention scoring for efficient pattern retrieval.

Provides model-agnostic prompt analysis and optimization via 8 research-backed patterns and a 3-step flow to enhance AI model performance.