Eric Martin
AI Systems Architect
Profile
AI Systems Architect bringing 19 years of operational discipline to AI infrastructure. Currently architecting the AI-augmented operations layer behind tax-efficiency and asset-protection planning at OCRC, where I manage active client cases and build the versioned automation portfolio the firm runs on. Independently building sovereign AI infrastructure on private hardware, including a production SMART on FHIR integration that powers a working caretaker-side system for elder care. Previously founded and scaled a Chicago-based design-and-systems consultancy serving 1,300+ clients across 19 years, entirely through word-of-mouth referrals.
Currently: Exploring senior roles bridging AI operations and regulated industries: healthcare systems, healthtech, financial services, insurance. Strong preference for organizations whose work touches eldercare, aging-in-place, or family caregiver coordination. Open to remote, hybrid, or Ohio-based.
Experience
- Architect and run the AI-augmented operations layer behind tax-efficiency and asset-protection planning. Manage active client cases from intake through delivery while building a versioned portfolio of Google Sheets templates, Apps Script automation suites, and AI-assisted workflows.
- Workflow Automation Suite: the firm's master client template through ~10 major versions in 8 months. Handles entity formation, document assembly, validation, and bulk operations across the active client roster.
- Document Heatmap: context-aware tracking and verification engine for multi-entity client engagements; identity-in-depth architecture, multi-tag support, automated batch operations.
- Bank Package Assembler: CRUT / Trust / LLC document packaging with execution-status enforcement, mandatory affidavits, and PDF watermarking. Client Onboarding automation generates full folder structures, entity skeletons, manifest documents, and secure API key transfer.
- Outcomes: ~95% reduction in new-client setup time; ~90% reduction in manual formatting time per document revision; eliminated formula-deletion and cross-row file collision errors at scale.
- Multi-LLM strategic deployment (GPT-5.2 Thinking, Gemini Pro 3, Claude) for code review, document drafting, and compliance synthesis, with different models picked for different tasks.
- Authored the OCRC Tools User Guide documenting system functions, authorization protocols, and output locations for seamless onboarding and knowledge transfer.
- Founded and scaled a design-and-systems consultancy serving 1,300+ clients across 19 years, scaling to six-figure annual revenue with $0 ad spend, 100% referral-driven growth.
- Delivered custom web applications, secure client portals, and workflow automation for verticals spanning nonprofit theater, commercial real estate, talent management, and healthcare.
- Key System: Built a HIPAA-grade physician portal for Specialdocs Consultants: login authentication, encrypted data storage, and private Q&A forum.
- Key System: Architected a three-division custom CRM for Grossman & Jack Talent, enabling independent pipeline management across TV/film, theater, and commercial operations.
- Managed digital systems, signage, and tenant programs for Class A office properties (including 111 S. Wacker Drive), coordinating deliverables across property managers, vendors, and JLL.
Independent Practice
- Built a working caretaker-side system on private hardware that handles parts of elder care that fall through the cracks of consumer apps: cross-system appointment coordination, medication context when prescriptions change, clinical changes translated into plain language, and the cognitive load of holding it all in one head.
- Production SMART on FHIR integration with a major EHR via OAuth 2.0 + PKCE under explicit patient authorization. Every external action passes a consent gate with a documented allowlist; full per-action audit log; PHI scoped to caretaker access with no leakage to non-medical lanes.
- Direction: making this kind of infrastructure available to family caregivers doing this work alone, a category of millions getting almost no help.
- Architected and deployed a persistent autonomous AI on private infrastructure with ~100 self-monitoring operational vertebrae handling daily operations, server health, security audits, and long-term memory consolidation.
- Designed a tiered intelligence layer: local LLM (Ollama/Llama) for low-cost reflex monitoring; cloud LLM (Claude/Gemini) for complex reasoning. Operational inference costs reduced by 97%.
- Built integrated scheduling, voice I/O, and cross-platform messaging: real-time AI-assisted communication and task management across Signal, calendar, email, and home automation.
- Operates under a Field Manual: internal governance with mechanical pre-commit enforcement, failure-class durability tracking, and per-vertebra cadence audits.