Consulting
AI Governance &
Agentic Systems Architecture
I build and audit governance infrastructure for teams running AI agents in production.
Working code, not slide decks. Stdlib-only Python, zero dependencies, battle-tested
across a 12-agent fleet.
Multi-Agent Orchestration
AI Governance & Compliance
EU AI Act Readiness
Platform / SRE
Claude / Anthropic
Delegation & Identity
Three engagement models, all backed by production-grade governance primitives.
Start with an assessment, scale to a retainer.
Targeted governance packages for specific regulatory and compliance needs.
Each package includes assessment, implementation, and documentation.
Governance
Agentic AI Governance
$20–40K
Governance architecture for autonomous AI agents: delegation chains,
identity management, human oversight, and multi-agent coordination protocols.
- Agent autonomy boundary definition
- Delegation token architecture (HMAC-signed)
- Kill switch and circuit breaker implementation
- Multi-agent coordination governance
- Singapore Agentic AI Framework alignment
Compliance
AI Employment Compliance
$8–15K / state
Bias audit, notice, and impact assessment compliance for AI hiring tools.
Covers NYC LL144, Illinois HB 3773, Colorado AI Act, and EEOC guidance.
- Independent bias audit coordination
- Disparate impact analysis by protected class
- Candidate notice and opt-out procedures
- Documentation and record retention
- Multi-state compliance mapping
Certification
ISO 42001 Certification
$25–50K
Full AI Management System (AIMS) implementation and certification readiness.
Scope definition through internal audit, ready for registrar assessment.
- AIMS scope and context definition
- AI risk assessment and treatment process
- Policy suite and procedure documentation
- Internal audit and management review
- Registrar selection and preparation
Framework
NIST AI RMF Implementation
$15–30K
Implement the four NIST AI RMF functions (Govern, Map, Measure, Manage)
with production-ready governance primitives.
- AI risk governance structure design
- Risk categorization and mapping
- Trustworthiness measurement framework
- Control implementation and monitoring
- Federal procurement alignment
Compliance
EU AI Act Conformity
$30–60K
High-risk AI system conformity assessment preparation. Technical
documentation, quality management, and post-market monitoring.
- Risk classification and scoping
- Technical documentation package
- Quality management system for AI
- Post-market monitoring plan
- Notified body assessment preparation
Sector
Financial AI Governance
$25–50K / jurisdiction
AI model risk management for financial services. Covers SEC disclosure,
UK FCA guidance, and OSFI E-23 requirements.
- AI model risk management framework
- Algorithmic trading oversight
- Consumer outcome testing
- Board-level AI governance structure
- Regulatory disclosure preparation
Strategy
Cross-Framework Mapping
$15–25K
Map your existing governance controls across all applicable frameworks.
Build once, comply everywhere. 70-90% cost savings vs. separate programs.
- Current-state governance inventory
- Multi-framework gap analysis
- Unified architecture design
- Prioritized implementation roadmap
- ROI quantification and business case
Review
AI Claims & Disclosure Review
$5–10K
Audit AI marketing claims and disclosure practices against FTC Section 5,
SEC requirements, and California AB 2013.
- AI capability claims audit
- Training data disclosure review
- Marketing materials compliance check
- SEC filing AI disclosure preparation
- Remediation recommendations
01
Discovery Call
30-minute call to understand your agent architecture and compliance needs. Free, no commitment.
02
Scope & Proposal
Written proposal within 48 hours. Fixed price, clear deliverables, defined timeline.
03
Execute
I work in your codebase or deliver standalone. Daily async updates, weekly sync calls.
04
Deliver & Handoff
Written report, working code, documentation. Your team owns everything.
About
Reuben Bowlby
Founding Architect of Founder Mode, an AI orchestration platform running 12 concurrent
Claude terminals with contract-driven cost governance and agent-first security.
I built HUMMBL's governance stack from scratch: delegation tokens (HMAC-SHA256 signed),
append-only governance bus, compliance mapper (SOC2 + GDPR + OWASP + EU AI Act),
and quality scoring via Arbiter. All stdlib-only Python, zero third-party runtime dependencies.
Creator of Base120,
a 780-model library (120 Base120 taxonomy + 660 extended models) for AI agent reasoning.
Published hummbl-governance on PyPI.
By the Numbers
Production Evidence