Direct answers before the proposal.

Foundation ($25,000) validates one use case and produces a roadmap. Workflow Launch ($75,000) releases a focused automation lane for one team. AI Operations ($100,000) adds a broader knowledge, agent, and workflow layer across more than one system. Operating Layer ($250,000) is the complete six-month local-first deployment. Enterprise is custom-scoped for programs beyond that reference size or complexity.

Two Mac Studio compute-node allowances; approved accessories; locally installed model runtime and designated models; business knowledge retrieval; a custom-configured Hermes agent; CRM and workflow automation; agreed marketing, sales, customer-service, and content workflows; a one-year third-party software/API allowance where listed; installation; six months of training; six months of maintenance and support; governance; testing; and acceptance documentation. The executed SOW and schedules control.

No. Hardware is included only when the package and final bill of materials expressly list it. The $250,000 Operating Layer uses a two-node Mac Studio reference. Smaller packages do not automatically include hardware, and Enterprise infrastructure is custom-designed.

Its role, instructions, approved tools, permissions, knowledge sources, workflows, approval points, escalation rules, and evaluation cases are configured for the client. “Custom” does not mean DGM owns the underlying third-party models or open-source frameworks.

Ownership is item-specific. The client is expected to own purchased physical equipment and its own data. Third-party software, model weights, APIs, and frameworks remain governed by their licensors. Rights in DGM deliverables, pre-existing materials, configurations, and custom work must be stated in the executed agreement.

Not necessarily. Designated local-model and knowledge workloads can stay in the approved client environment. Connected CRM, email, advertising, or other API workflows may send data to named third parties. The final data-flow map discloses those paths.

Yes. AI output can be incomplete, inaccurate, biased, or inappropriate. DGM uses retrieval, evaluations, permissions, approval points, logging, and escalation to reduce risk, but does not guarantee perfect output.

No universal certification or compliance claim is made. The SOW can define specific controls and evidence, but legal and regulatory suitability depends on the client’s facts, industry, data, jurisdiction, and qualified review.

Not in the reference architecture. An air-gapped design has different integration, update, support, and operational requirements and would need separate engineering and scope.

Enterprise is for programs beyond the $250,000 reference deployment, such as multi-site, multi-entity, regulated, high-availability, or multi-department rollouts. Architecture, integrations, governance, rollout, support, and service levels can vary materially, so DGM prices that scope only after discovery.

No. The illustrative allocation is $36,000 equipment, $24,000 software/API allowance, $116,000 engineering and integration, $24,000 training, $30,000 support and maintenance, and $20,000 program and acceptance services.

No. Independent financing sources make their own decisions about eligibility, structure, financed categories, rate, term, payment, collateral, and documentation. DGM can provide accurate transaction documents but cannot guarantee an outcome.

Published package amounts are before applicable taxes. Final documents state treatment of shipping, taxes, usage overages, change orders, and costs outside included allowances.

Each package includes only the third-party products, terms, and allowances named in its final software schedule. Renewals are the client’s responsibility unless that schedule says otherwise. It records the account owner, renewal owner, term, allowance, and known usage limits.

No. Foundation is planned as a 4–6 week sprint, Workflow Launch as an 8–12 week implementation, AI Operations as a 3–4 month implementation, and Operating Layer as a six-month implementation. Enterprise timing is custom. Contracting, hardware availability, client access, data readiness, approvals, review delays, technical findings, and change requests can affect any schedule.

Support is package-specific: Foundation includes closeout and roadmap support, Workflow Launch includes a 60-day support window, AI Operations includes a 90-day support window, and Operating Layer includes six months of maintenance and support. Enterprise support and service levels are negotiated. The final schedule defines channels, hours, response targets, maintenance, exclusions, contacts, and change requests.

Against measurable technical and service criteria in the SOW: equipment inventory, configuration, agreed test cases, workflow behavior, documentation, training, and support handoff. Acceptance is not a guarantee of revenue, savings, ROI, or error-free AI output.

No. The booking page collects a preferred date, time, and time zone. DGM confirms that time or proposes an alternative separately. Any downloaded calendar file is labeled tentative and does not create a reservation.

The site validates the request using a same-origin intake endpoint. When DGM connects its approved workflow webhook, validated requests can be delivered there. Until then, the site opens a prepared email to Yeshua@kairossystems.ai for the visitor to review and send. The current process is described in the Privacy Policy.