Advisory & consulting

Select engagements in agent infrastructure, deterministic state, and machine-native commerce.

Context

I take on a small number of consulting engagements in areas where I have deep, current, hands-on expertise. Each engagement is chosen for technical overlap with the infrastructure problems I work on daily.

My background spans nearly two decades across consumer web, developer platforms, and crypto infrastructure: writing and shipping at TechCrunch, co-founding Plancast, leading user experience at Hiro for the Stacks blockchain, and running Leather at Trust Machines. I now build Neotoma, a structured memory layer for AI agents, and operate with a full agentic stack where AI agents function as collaborators rather than tools. You can see the full arc on my timeline.

Focus areas

Agent systems & AI infrastructure

Architecture and reliability for AI systems that take actions, maintain state, and evolve over time.

  • Multi-step autonomous workflows and orchestration
  • Tool-using LLM systems and agent runtimes
  • Evaluation pipelines and observability
  • Debugging reliability problems in agent execution
  • Data pipelines feeding agent workflows

Deterministic state & knowledge architecture

Systems where correctness, traceability, and reproducibility of state are critical.

  • Event-sourced architectures and versioned data
  • Knowledge graphs and entity resolution
  • Provenance tracking and audit systems
  • Long-running workflow state management

Machine-native commerce

Programmable payment infrastructure for autonomous software agents.

  • Agent-controlled wallets and custody models
  • Agent-to-agent payment architectures
  • Programmable settlement and API-native payment flows
  • Infrastructure enabling autonomous services to transact

Types of engagements

Architecture review

Focused analysis of system architecture, identifying structural weaknesses, reliability risks, and opportunities.

Technical advisory

Ongoing advisory providing architectural guidance and strategic technical input.

System debugging & reliability

Diagnosing complex failures in AI systems, automation pipelines, or distributed systems.

Platform & infrastructure design

Advising on the architecture of new platforms involving agents, workflows, or programmable infrastructure.

Product architecture & developer experience

Advising on how product design maps to system architecture for developer-facing AI tools. API surface design, developer onboarding, and adoption strategy for infrastructure products.

Best fit

  • Technically ambitious systems at the infrastructure level
  • Teams with strong engineering who need architectural or product-architecture perspective
  • Problems involving state evolution, reproducibility, or complex automation
  • Willingness to explore unconventional architectural approaches

Working structure

  • 5–10 hours per week, defined scope and timeline
  • 2–8 weeks for architecture engagements
  • 1–3 months for advisory retainers
  • Fixed-scope projects, architecture sprints, or monthly retainers
  • No embedded team participation or day-to-day engineering roles

Open-source commitment

Neotoma is developed as open infrastructure. Engagements must allow continued development of Neotoma, reuse of general architectural ideas, and publication of open-source work.

Client-specific work stays confidential. Engagements requiring exclusivity over core architectural concepts or restricting open-source development are not accepted.

Get in touch

If you think your project may be a good fit, I’d be happy to discuss it.

Email me and please include:

  • A short description of the system you are building
  • The technical challenges you are encountering
  • The type of engagement you have in mind
  • Expected timeline