3. Data, Science & AI

Offers industry-proven design patterns for AI agent architectures, context management, and sub-agent implementations based on production systems.

Designs persistent memory architectures for AI agents using Vector RAG, GraphRAG, and Temporal Knowledge Graphs.

Detects errors in AI-generated output before finalization using a self-critique loop with 'I'm in danger' detection.

Executes HEC-HMS hydrologic simulations with parallel processing, version management (3.x/4.x), and Jython script coordination.

Updates HEC-HMS meteorologic model files for precipitation, gage assignments, evapotranspiration, and Atlas 14 frequency storms in hydrologic modeling workflows.

Statistical analysis toolkit for hypothesis testing, regression, correlation, and generating APA-compliant reports for academic research.

Applies permanent fixes for critical AI model failures including hallucinations and instruction drift through forensic investigation.

Provides an operational manual for optimizing context usage, troubleshooting performance, and understanding the mechanics of the Claude AI runtime.

Generates structured, multi-stage prompts for Claude-to-Claude pipelines enabling complex AI workflows with research, planning, and execution stages.

Converts GitHub repositories into structured plain-text digests optimized for LLM-based codebase analysis and AI ingestion.

Provides complexity-based standards, patterns, and quality criteria for designing, optimizing, and auditing AI prompts, system instructions, and multi-stage chains.

Defines and enforces behavioral standards for autonomous agents, including Uninterrupted Flow, Self-Verification, Auth-Gates, and Handoff protocols.

Provides a unified interface for AI agent context optimization including compression, degradation detection, and KV-cache management.

Provides foundational SPARQL patterns for querying RDF data and knowledge graphs, enabling semantic web data retrieval and manipulation.

Configures Home Assistant energy monitoring with dashboards, solar/grid tracking, and consumption analysis for data-driven energy insights.

Optimizes Frigate NVR configuration for object detection, recording, zones, and hardware acceleration via YAML, including troubleshooting and Home Assistant integration.

Analyzes database schemas and query performance to identify bottlenecks and suggest optimizations.

Analyzes database schemas and relationships to identify optimization opportunities and improve query performance.

Provides zero-cost conversion tracking for Astro.js lead gen sites using GA4, Meta CAPI, and Google Ads with consent management and Safari ITP bypass.

Ingests career data from Obsidian, CSV, and markdown into validated YAML/Markdown for CV portfolio systems.

Enables PACT agents to persistently store and semantically search context, decisions, and lessons across sessions for AI-driven memory management.

Sets up analytics and conversion tracking for lead generation websites using GTM, GA4, and Astro framework integration.

Analyzes FDA adverse event data to detect drug safety signals, compare safety profiles, and provide insights on side effects, recalls, and special population safety.

Supports clinical trial safety planning using FDA adverse event data, covering safety monitoring, protocol sections, and special population considerations.