3. Data, Science & AI

Optimizes PostgreSQL, MySQL, and MongoDB queries via EXPLAIN analysis, index recommendations, and N+1 query detection.

Automates Jupyter notebook cleanup: removes empty cells, structures code, extracts functions, and generates documentation for improved readability.

Guides end-to-end development of stateful, multi-agent AI applications using LangGraph, including StateGraph design, RAG integration, and production deployment.

Orchestrates parallel AI agents to evaluate and compare outputs for complex tasks, selecting the optimal solution through competitive benchmarking.

Enriches creator profiles by integrating Talent API data to improve discovery and ranking algorithms.

Fetches real-time website traffic analytics data from Plausible for data analysis and reporting.

Transforms team photos into professional startup portraits using AI image-to-image processing.

Extracts articles and content from websites using FireCrawl API for data processing and analysis.

Generates blog featured images via fal.ai Nano Banana Pro API, leveraging AI for on-demand visual content creation.

Discovers Instagram influencers through keyword-based searches using the Bright Data API.

Retrieves user metrics (total, active, new users) from Clerk for data analysis and monitoring purposes.

Enables keyword-based search for TikTok influencers using Bright Data API, delivering data for marketing and social media analysis.

Scrapes and extracts structured content from web pages via Firecrawl API for data processing and analysis.

Enables structured storage and management of influencer data within a Notion database.

Converts scanned PDFs and image-based documents to clean markdown using OCR, outputting a markdown file adjacent to the original.

Manages local project-specific MCP (Model Context Protocol) servers, enabling addition or removal of servers like context7 or chrome-devtools.

Provides vector search alternatives for SQLite projects, including SQLite VSS, external vector DB integration, and RAG pipeline implementation.

Provides standardized guidelines for AI agent development, covering frontmatter, naming conventions, tool access, model selection, and Bash file operations in .opencode/ folders.

Optimizes AI model context windows by measuring tokens, prioritizing content, compressing data, and validating outputs within token constraints.

Models real experts' thinking patterns and expertise into AI agent personas for sophisticated behavior in complex systems.

Guides selection of Anthropic AI models (Sonnet, Haiku) for agents based on task complexity, reasoning needs, and performance requirements.

Systematizes database normalization requirements, normal form judgment, and trade-offs between normalization levels (1NF-5NF) and performance for relational schema design and review.

Enables design and state management of multi-turn conversations, ensuring consistent dialogue experiences through context and turn management.

Guides selection between Anthropic's Sonnet and Haiku models based on task complexity, reasoning needs, and performance for agent implementations.