Imagination and discovery
Identify repeated problems, hidden context, trust concerns, and workflow moments where AI could create surplus value with the right operating context.
/~
A forward-deployed platform that makes your work observable, your decisions grounded, and your actions governed around how your team actually works.
/ Use cases
Each use case begins with a focused operating area where the work already happens, then helps the team see what needs attention, what is missing, and what should move next.
The examples here show where that operating layer can be applied across different teams, workflows, and review rhythms.
For firms where client work depends on evidence, deadlines, judgment, and review. The system can be configured to keep matter context together: what changed, what is missing, which documents support the current position, and what needs approval before anything moves forward.
Read use caseFor teams managing receipts, approvals, renewals, spend, owners, and review cycles. This can be configured to make the control context visible: what needs attention, what evidence exists, who owns the decision, and what should be escalated before action.
Read use caseFor studios where work moves through references, options, client conversations, consultant input, and review notes. For this kind of practice, the system can keep the project story together: why decisions were made, what changed after critique, and what needs preparing before the next review.
Read use caseFor teams working across site evidence, drawings, inspections, constraints, contractors, and formal approvals. In this setting, it can keep the technical record clear: what changed, what supports it, which issues remain open, and who owns the next step.
Read use caseFor teams where continuity, sensitivity, and handover context matter. The system can be configured to prepare reviews and follow-ups from the approved record while keeping sensitive decisions and client-facing actions behind human judgment.
Read use caseFor galleries, hospitality, media, partnerships, and founder-led teams where trust lives across conversations and follow-through. For relationship-led work, it can keep relationship memory visible: what was promised, who owns the next step, what context matters, and what needs a careful follow-up.
Read use case/ How It Works
We embed an engineer with your team to understand the systems, workflows, decisions, and review habits already in motion, then keep working alongside you as the operating layer is deployed, adopted, and expanded across the business.
Identify repeated problems, hidden context, trust concerns, and workflow moments where AI could create surplus value with the right operating context.
Define the operating area: the source systems, evidence, language, approvals, data boundaries, and review rhythm needed to make the workflow loop useful and visible.
Build a prototype layer around the chosen workflow and test it with the team, using real materials, real decisions, and real review conditions to understand what the system needs to support.
Turn the proven prototype into an adopted workflow, establishing the ownership, controls, feedback rhythms, and operating habits needed for it to run reliably and shape the next loop.
/ Operating Platform
Starting from a three-layer model of professional work, this gives every deployment a common structure while still allowing the system to be configured around the language, workflows, systems, and constraints of each practice.
The shared model of how the practice thinks: principles, terminology, goals, standards, and the reasons decisions are made a certain way.
The repeated operating rhythm of the practice: reviews, handovers, follow-ups, ownership, deadlines, and moments where judgment is required.
The resource context behind the practice: subscriptions, vendors, tools, and commitments understood by what they support and who owns them.
/ Orchestration layer
Actor is a minimal-setup custom agent harness connected directly to your operating twin. It can answer questions, complete bounded tasks, set up workflows, prepare summaries and follow-ups, and queue actions for human review before anything important is written or sent.

/ Trust and control
Proximity is built for environments where expertise, accountability, and professional discretion matter. The system can prepare context, surface patterns, draft next steps, and organize review queues, but sensitive decisions and external actions remain governed by the people responsible for the work.

/ Deployment and data control
Proximity does not require every practice to adopt the same hosting, model, or data posture. Each deployment can be configured around the organization's infrastructure requirements, approved model providers, existing subscriptions, and internal governance policies.
Run Proximity in a managed tenant or self-host it within the organization's own environment.
Use approved model providers and existing subscriptions instead of being locked into a single model stack.
Keep operating context, permissions, and access scoped to the teams, workflows, tenants, and environments defined for the deployment.
/ Start
Begin with a focused review rhythm, workflow, or team where better operating context would immediately change the quality of preparation and judgment.