3. Data, Science & AI

Optimizes and structures LLM prompts for clarity, consistency, and task-specific performance across various models.

Builds and troubleshoots multi-model AI orchestrators with OpenRouter and Aider, resolving API errors, large file handling, and JSON output issues.

Manages Oracle database operations, including creation and administration of database links for cross-database data integration.

Manages and adds model evaluation results to Hugging Face model cards, supporting extraction from READMEs, API imports, and custom evaluations with vLLM/lighteval.

Provides best practices for comprehensive Jupyter notebook data analyses with statistical rigor, outlier handling, and publication-quality visualizations.

Manages Elasticsearch indices with CRUD operations, aggregations, and mappings for efficient data search and analytics.

Analyzes fuel business data via SQLite and Laravel to generate KPIs, dashboard proposals, and actionable operational insights for leadership.

Enables creation, management, and use of custom Agent Skills for Claude, covering architecture and best practices.

Manages PostgreSQL database operations including backups, transactions, user management, window functions, and materialized views.

Specializes in geospatial data engineering and cartographic visualization for spatial analysis and data-driven storytelling.

Creates structured agent prompts with behavior rules, intent mappings, and anti-hallucination policies for reliable AI agent operation.

Creates, edits, and analyzes Excel spreadsheets with formulas, formatting, and data visualization for financial models and reporting.

Extracts structured data from websites via scraping and browser automation, bypassing blocks and CAPTCHAs for live content and product data.

Provides quick reference for configuring AI tools including Google Gemini and OpenAI Codex.

An AI runtime toolkit integrating internal and external AI tools with discovery, execution, and configuration capabilities.

Core identity and command system for an AI assistant enabling exploration, learning, planning, and execution via /runtime.* commands with constitutional governance and memory management.

Provides context-aware, personalized AI conversations by leveraging user profiles and vectorized notes for memory and reference in ongoing dialogues.

Enforces standardized SQL schema conventions for PSRDatabase to ensure consistency and maintainability in database design.

An AI memory system enabling SQL-style event queries, timeline management, and intelligent memory retrieval for project history tracking and experience transfer.

Assists in constructing and interpreting Plausible Analytics API queries for website traffic metrics, dimensions, and time ranges.

Extracts structured data, text spans, metadata, and images from PDF pages, creating persistent artifacts for data pipelines and downstream analysis.

Builds containerized neural memory systems with test-time learning capabilities using Titans/TTT architecture for dynamic memory modules.

Transforms PDF pages into semantic HTML using AI, utilizing PNG, parsed text, and ASCII preview inputs for structural accuracy.

Enables building machine learning and AI applications in Rust, supporting model training, inference, and neural networks with specialized libraries.