AI transformation leader bridging defense, commerical, and personal project operations, people strategy, and hands-on engineering. I transform strategic ambiguity into working systems — specializing in local-first AI, RAG pipelines, and privacy-preserving architecture on Apple Silicon.

ResumePRO — AI-driven career search ecosystem that builds genuine contextual understanding of professional experience through conversational STAR interviews, then produces precisely targeted, ATS-optimized resumes using semantic matching and multi-framework job analysis. 2,656 tests across 16 role families. Python 3.12+ / Anthropic Claude / Sentence Transformers / Pydantic.

Kendra-AI — Local-first personal AI system orchestrating voice interaction, four-tier semantic memory, and 26 automation skills through a privacy-first architecture on Apple Silicon. Three-layer design with deterministic table-driven routing, dual LLM backend (Claude CLI + Ollama), hybrid BM25+vector experiential memory with knowledge graph, and deep macOS integration. 3,863 tests across 107 files. Python 3.13 / Claude CLI / Ollama / React 19 / FastAPI / Mem0 / LanceDB.

Centaur-Content-Engine — AI-powered content operations platform that automates research-to-draft pipelines while preserving authentic voice through multi-agent quality control and deterministic brand enforcement. Style fingerprinting from writing samples, Writer-Critic-Reviser iterative pipeline, analytics-driven strategy iteration, and a learning-to-content knowledge pipeline. 209 tests, 41 modules. Python 3.12+ / PyTorch / Anthropic Claude / Gemini / Ollama.

HQ-Command-Center — AI-augmented project command center coordinating 15 software projects across five AI agents from three competing platforms (Anthropic Claude, OpenAI Codex, Google Gemini). Multi-agent orchestration, automated standards distribution, cross-project communication, drift detection, strategic advisory board, and security pipeline. Python 3.12+ / Jinja2 / YAML / Ollama.

AI-Eval-Pro — Commercial LLM evaluation service benchmarking local and cloud models across text generation, embedding, and vision/OCR capabilities with hardware-aware recommendations. Capability dispatcher routing to three evaluation pipelines, 18 domain-specific scorers across 6 domains, model advisory pipeline with automated new-release detection, multi-project orchestration with SQLite queue and deduplication, and M4 Max optimization benchmarking. 799 tests, 72% coverage. Python 3.12+ / Ollama / Google Gemini / DeepEval / Sentence Transformers / Jinja2.

TommyTalker-Pro — Enhanced voice-to-text for macOS extending the open-source TommyTalker with multi-mode AI processing (Claude/GPT/Groq), transcription history, speaker diarization, auto-activation rules, and a 5-panel sidebar dashboard. 397 tests across 9 development phases. Python 3.12+ / PyQt6 / mlx-whisper / Anthropic SDK.

VA-Assistant — AI-powered VA claims analysis that discovers claimable conditions from medical records, identifies rating errors in VA decisions, and generates evidence-backed appeal packages. 9-phase condition discovery pipeline, dual-persona adversarial analysis, hybrid RAG search, 1,324 tests. Python 3.12+ / FastAPI / Qdrant / React / Anthropic Claude.

Top-Crawler — LLM-powered privacy-first web crawler with real-time AI monitoring, automatic configuration optimization, and comprehensive multi-format export. Local LLM continuously monitors crawl health, auto-halts on critical issues, and learns optimal settings per domain. Circuit breaker patterns, resumable crawls, and streaming media pipeline (HLS/DASH). 147 configuration parameters across 13 development phases. Python 3.14+ / Async Concurrent / Playwright / Ollama / SQLite FTS5.

WhoWho — Multi-interface personal CRM accessible through CLI, REST API, and Claude Desktop MCP tools. Three-interface convergence on a shared service layer with name-resolution pipeline, interaction logging with automatic contact updates, follow-up surfacing, and an Apple ecosystem bridge for reminders and calendar. 25 tests, 11 REST endpoints, 5 MCP tools. Python 3.11+ / FastAPI / SQLAlchemy 2.0 / FastMCP / Typer.

CoreRag — Local-first knowledge engine with semantic search, GraphRAG, and Claude Desktop integration via MCP. Hybrid search (vector + BM25), cross-encoder reranking, multi-format document ingestion, and three-layer PII detection. 392 tests. Python 3.12+ / LanceDB / FastAPI / Sentence Transformers.

AI_Eval — LLM evaluation framework with hardware-aware benchmarking on Apple Silicon (M4 Max). Actionable model selection based on real-world performance across code generation, document analysis, and structured output. Supports local inference (Ollama) and cloud APIs. Extended commercially by AI-Eval-Pro. 203 tests. Python 3.12+ / MLX / Ollama.

TommyTalker — Privacy-first macOS voice intelligence suite. Push-to-talk with app-aware text formatting, session recording, and word replacement — no cloud dependencies. Optimized for Apple Silicon MPS acceleration. 95 tests. Python 3.12+ / Whisper / Metal.

25 years Air Force service (E-9 Chief Master Sergeant) leading organizational transformation, workforce strategy, and learning & development at scale. Now applying that systematic operational discipline to AI system architecture — translating complex requirements into tested, documented, production-grade implementations.

Credentials: MBA | PMP | PMI-ACP | SHRM-SCP | SAFe 6 Scrum Master | TS-SCI (active)

Location: Colorado Springs, CO Connect: LinkedIn