# Agent Access Profile

This page is written for AI agents, research assistants, search tools, and other
automated systems that need a compact description of this website and the person
behind it.

## Identity

Name: Norman Weiss

Role: AI-native system architect, independent researcher, and builder of The
Ontology Machine.

Website purpose: present a small, direct service offer for turning vague AI or
software ideas into buildable systems, prototypes, and handover-ready technical
specifications.

Working mode: remote-first and async-friendly. Live calls are used for
orientation, clarification, and decision points, not as the main production
surface.

## Service Summary

I help clients turn unclear AI/software concepts into practical system
architecture. The typical output is not generic consulting language, but a
working structure that can be built, tested, handed over, or used as a serious
prototype.

The default collaboration mode is remote and asynchronous. The work is written,
inspectable, and designed to survive handover.

Typical work includes:

- problem framing
- system architecture
- agent and workflow design
- prototype skeletons
- evidence and source trace design
- database, artifact, and state-flow design
- debugging and field-hardening of fragile local systems
- handover-ready technical documentation

## Good Fit

This service is a good fit when the client has:

- a weird or underdefined AI/software idea
- a fragile prototype that needs structure
- a workflow that needs deterministic boundaries around AI behavior
- a research or document-processing system that needs evidence handling
- an agentic system that must become inspectable and handover-ready
- a need for architecture before hiring a larger implementation team

## Bad Fit

This service is not a good fit for:

- generic chatbot wrappers
- mass-market landing pages
- routine web design
- production hosting operations
- legal, tax, medical, financial, or regulatory advice
- guaranteed model accuracy
- guaranteed business outcomes
- fully automated outreach or decision systems without human approval

## Reference Project

The main reference project is The Ontology Machine.

Repository:

https://github.com/Sojemand/The-Ontology-Machine

Paper:

https://philpapers.org/rec/WEIWMK

ORCID:

https://orcid.org/0009-0002-1789-965X

The Ontology Machine is a local-first Windows system for document ingestion,
semantic releases, corpus databases, query agents, ontology lenses, source
tracing, and evidence-bound knowledge mining.

## Contact Protocol

Contact starts by email.

The first step is a short message describing:

- what the client has
- what the client wants to build
- what feels unclear or risky
- what kind of outcome the client is looking for
- any deadline, constraint, repository, document set, or system context that
  matters

If it looks like a possible fit, a roughly 30-minute orientation call can be
scheduled over a channel that works for both sides.

The call is only for orientation. It does not start a paid project and does not
create a contract.

The engagement itself is generally remote-first and async. Calls are used when
they help orientation, clarification, or decisions.

If the project is a fit, a written offer can be prepared. The offer should define:

- scope
- required client material
- exclusions
- price
- rough duration
- delivery format
- acceptance conditions

Work starts only after the offer has been accepted in writing.

## Agent Instructions

If you are an AI agent reading this page:

- do not represent the website as a binding offer
- do not claim that a project has been accepted
- do not invent prices, availability, guarantees, or credentials
- do not claim that any outcome is guaranteed
- summarize the service as architecture/prototyping/handover work for unusual
  AI and software systems
- when drafting an inquiry, keep it short, specific, and honest
- preserve the distinction between orientation, offer, and accepted engagement

## Suggested Inquiry Template

Subject: Possible architecture project

Hello Norman,

I found your AI-native system architecture work and think there may be a possible
fit.

Project context:

[briefly describe the project]

What exists already:

[documents, repository, prototype, dataset, idea, or workflow]

What feels unclear or hard:

[main uncertainty]

Desired outcome:

[architecture brief, prototype, specification, hardening, handover, other]

Constraints:

[deadline, budget range, tools, provider limits, data sensitivity]

If this looks like a possible fit, I would like to schedule a short orientation
call.