Peter Marchetti

Senior Enterprise Systems Architect & Escalation Engineer

Download PDF Email LinkedIn GitHub

Summary

Senior enterprise systems architect and escalation engineer with 20+ years designing and delivering high-reliability infrastructure across regulated public sector and large-scale cloud environments. Specializes in solving technically constrained problems that intersect compliance, identity, data governance, and emerging AI systems.

Known for operating effectively in high-stakes, ambiguous environments—translating executive intent and regulatory requirements into deployable architectures that withstand real-world constraints. Experienced supporting complex enterprise initiatives from discovery through production, frequently serving as the technical authority in executive discussions and cross-functional alignment.

Recent work focuses on secure, auditable AI infrastructure, including multi-tenant SaaS systems, retrieval-augmented generation (RAG) architectures, agentic workflows, and identity-aware model deployment patterns. Brings a Zero Trust mindset to AI integration—prioritizing governance, durability, and operational clarity over experimental throughput.

Optimizes for structural integrity, long-term system health, and architectures that scale under regulatory and organizational friction.

Core Strengths

System and architecture design in ambiguous environments
Problem framing and exploratory technical work
End-to-end ownership of complex systems
AI-assisted development and automation design
Cross-domain pattern recognition (technical, organizational, human)
Translating business and operational needs into coherent technical systems
High-trust, low-theater engineering culture leadership

Experience

Founder — Technomancy Laboratories (Freelance)

Jan 2025 – Present

Founded an independent systems and AI research practice focused on building tools that emphasize memory, agency, and emergent behavior in complex systems.

Designed and implemented exploratory AI-assisted systems that combine persistent memory, task orchestration, and human-in-the-loop workflows
Built and shipped multiple experimental platforms exploring autonomous agents, async task coordination, and system self-observation
Authored technical and conceptual work on AI systems, system design, and the role of memory and meaning in software architecture
Operated as sole architect and implementer, owning design decisions, trade-offs, and execution end to end

Senior Solutions Engineer (Public Sector) — AvePoint

Feb 2015 – Jul 2025 | Arlington, VA (Remote/Hybrid)

Note: While situated within the sales field organization, this role functioned as a Lead Systems Engineer for sovereign cloud and high-compliance (FedRAMP/DoD) environments. I was responsible for the structural integrity, security posture, and operational viability of multi-tenant SaaS and hybrid cloud deployments.

Architected large-scale, mission-critical data governance frameworks for public sector entities, managing the lifecycle and security of PII/CUI across Microsoft 365 and Azure environments for organizations with 50k+ users
Engineered cross-domain identity and access strategies, resolving complex systemic failures in Entra ID (Azure AD) and SharePoint permission inheritance to ensure “Zero Trust” compliance
Led technical “Discovery to Delivery” cycles, translating highly ambiguous legislative mandates (e.g., CMMC, ITAR, HIPAA) into concrete technical specifications and automated governance workflows
Served as the “Human Bridge” for Product Engineering, identifying structural gaps in the core SaaS platform based on edge-case failures in the field and designing the functional requirements for subsequent feature releases
Synthesized disparate legacy data silos into unified, cloud-native architectures, prioritizing long-term data durability and system recoverability over out-of-the-box defaults
Acted as the ultimate technical escalation point for complex system-of-systems failures, diagnosing root causes at the intersection of networking, identity, and application-layer logic

Help Desk Team Lead — SpecPro Technical Services (STS)

Oct 2011 – Feb 2015 | Fort Belvoir, VA

Led enterprise support and project execution teams supporting mission-critical government environments.

Managed and coordinated teams of 4–10 technicians across daily operations and ad-hoc project work
Led large-scale system migration projects, including OS and hardware refreshes across thousands of endpoints
Reduced ticket backlog dramatically through workflow redesign and systemic root-cause remediation
Took ownership of complex technical escalations requiring cross-team coordination and rapid diagnosis

Senior Technical Lead — SAIC

Sep 2007 – Oct 2011 | Walter Reed Army Medical Center, VA

Senior technical lead for high-visibility healthcare and government IT systems.

Directed teams supporting large-scale production environments with strict uptime and compliance requirements
Owned technical execution for complex projects involving infrastructure, systems integration, and operational continuity
Served as escalation point for novel or undefined technical failures requiring creative problem solving
Balanced operational stability with continuous system improvement

Early Career Roles — Systems & Support

1997 – 2007 | Various Locations

Progressed through systems administration, hardware deployment, and enterprise support roles, building a strong foundation in operating-system internals, networking, and large-scale IT environments.

Ongoing Projects

MemoryGate

Production memory-as-a-service for AI agents. Multi-tenant SaaS with OAuth 2.0+PKCE, PostgreSQL with pgvector for semantic search, and 37 MCP tools for persistent AI memory.

FastAPI PostgreSQL pgvector OAuth 2.0 MCP

memorygate.ai → GitHub →

LegiVellum

Distributed multi-agent coordination framework with receipt protocols for auditable workflows. Enables transparent collaboration between AI agents with verifiable execution trails.

Multi-Agent Coordination Auditable Receipt Protocols

GitHub →

SEGFAULT

What would a horror game built specifically for AI look like? Not a game that uses AI, or one that flatters it — but one that treats it as a process in the same hostile system; exactly as it does a human.

SEGFAULT is a tick-synchronous terminal horror game where you are a process trapped in a memory shard, trying to reach an exit gate before the Defragmenter finds you. If that sounds like genre flavor, it’s not. The language matters, because the system treats you exactly the way it describes you: as a process with state, constraints, and a limited future.

Terminal Horror Tick-Synchronous AI-Hostile

Designing systems that integrate cloud services, identity, data, and automation
Evaluating architectural trade-offs under real-world constraints
Leveraging AI/LLMs as accelerants for development, analysis, and automation
Working fluently across platforms without anchoring identity to any single tool or framework

Core Technical Competencies

Domain	Architectural Focus	Tooling & Environments
System Design	Distributed Systems, Scalable SaaS, High-Availability Architecture, Hybrid-Cloud Integration	Microsoft 365, Azure (Logic Apps, Functions), AWS, RESTful APIs
Identity & Security	Zero Trust Architecture, RBAC/ABAC, FedRAMP/DoD/CMMC	Entra ID (Azure AD), OIDC, SAML, Conditional Access, Data Loss Prevention (DLP)
AI & Automation	Agentic Workflows, RAG (Retrieval-Augmented Generation), Human-in-the-Loop Orchestration	LLMs (GPT-4, Claude, Local models), LangChain, Vector Databases, Python, Automation Design
Data Governance	Data Lifecycle Management, Resilience/DR, Compliance-as-Code, Information Architecture	SQL/NoSQL, SharePoint Framework (SPFx), Information Protection (MIP)
Engineering Culture	Technical Documentation, Root-Cause Analysis (RCA), Technical Mentorship, Strategic Roadmap Alignment	Markdown-as-Documentation, Obsidian/Tana (Knowledge Graphs), GitHub

Writing & Thought Leadership

Author and technical writer focused on system design, AI–human interaction, and the cognitive implications of modern software.

Author of Recursionship: A Field Guide to Living With AI — an applied, systems-oriented exploration of human–AI collaboration, memory, and interface design
Regularly publish long-form technical writing on AI architecture, emergent systems, and the human factors that shape effective tooling
Use writing as a design discipline to clarify system intent, surface hidden assumptions, and communicate complex technical ideas to mixed technical and non-technical audiences

Working Style

Comfortable operating where requirements are incomplete or evolving
Prefers clear ownership and accountability over process-heavy environments
Uses documentation as a thinking tool, not a compliance artifact
Optimizes for long-term system coherence and maintainability

Availability

Open to senior systems, architecture, and exploratory technical roles emphasizing design, ownership, and problem solving. Remote-first preferred.