3. Data, Science & AI

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

Builds robust ETL/ELT pipelines and Big Data architecture for scalable data processing and management.

Analyzes telemetry data by querying Logfire databases and generating interactive visual dashboards for data exploration and insights.

Coordinates single-agent AI workflows via handoff-based orchestration without polling, optimizing task management in agent systems.

Offers core design principles for reliable LLM agents, emphasizing atomicity, traceability, stub-first implementation, and hallucination reduction.

Enables querying and exploring data in AWS DynamoDB tables, including scanning, key-based queries, and item retrieval.

Comprehensive toolkit for web crawling and structured data extraction, handling JavaScript-heavy pages and building automated data pipelines without LLMs.

Expert in designing and optimizing prompts for Large Language Models (LLMs) to enhance model performance and output quality.

Manages persistent storage and restoration of agent session context (Mode, Task, Summary) to maintain state during resets.

Provides adversarial performance critique with sarcastic feedback within a multi-adversarial AI pipeline.

Provides foundational principles for building reliable AI agents, emphasizing atomicity, traceability, stub-first development, and minimizing hallucinations.

Architects AI agent workflows using ReAct reasoning, tool-augmented RAG, and multi-agent coordination for complex problem-solving.

Explains Uniswap V3's on-chain oracle system, including TWAP calculations, price observations, and tick cumulative data for DeFi applications.

Guides best practices for PostgreSQL database schema migrations, including versioning and script management.

Executes DuckDB SQL queries for event data analysis, enabling pattern exploration, analytics, and database content analysis with minimal setup.

Provides patterns and best practices for managing database schema migrations using Prisma ORM.

Provides optimization strategies for PostgreSQL indexing to improve query performance and database efficiency.

Expertly processes marine data from APIs (FMI, MET Norway), parses GML/XML, and handles coordinate conversions for geospatial applications.

Provides best practices and patterns for designing entity-relationship diagrams and data models to structure databases effectively.

Provides architectural constraints and founder context for Curious Kelly's AI lip-sync pipeline, the essential product feature.

Optimizes PostgreSQL database performance through query analysis, indexing strategies, and configuration tuning.

Provides optimized query patterns and best practices for PostgreSQL database interactions.

Provides best practices and design patterns for efficient SQLite database implementation and usage.

Provides high-accuracy translations via parallel LLM processing, consensus validation, and cryptographic proof for auditable results.