A model for your model
Proxa is a persistent workspace for your team and Claude that works with the files and tools you already use.
Definitions
What is a model?
A model is a simplified representation of a subject. It extends a knowledge base by adding context, relationships, dependencies, assumptions, and logic — so the subject can be understood, analyzed, and tested, not just searched or retrieved.
The idea shows up everywhere
Function
Finance, Legal, Sales, Product, etc. A recurring area of responsibility, expertise, or activity.
Project
Construction, Diligence, Research, Product Launch, etc. A defined body of work organized around an objective.
Asset
Buildings, Wells, Vehicles, Investments, etc. A resource that is managed, analyzed, or operated over time.
Organization
A higher-order Model composed of connected Function, Project, and Asset sub-models.
“A model is a map - not the territory”
Like a map - a model is reductionist by design — its value comes from what it leaves out. The objective is not to centralize everything an organization knows; a copy of everything is just the territory again, and no more navigable. The discipline is choosing what the work actually depends on, encoding that well, and letting the documents, decks, and reports draw from it rather than each holding their own version.
HOW PROXA WORKS
A dynamic model - not a folder of files
Proxa connects the work you already do to a model that keeps its structure, context, and next steps in view.
01
Like 100 tabs in Excel
Proxa brings the data, relationships, and logic needed for the Model into one connected structure, while the files and tools you already use stay in place. It eliminates silos to create a single version of truth for the team.
Dive deeper →
Data
The facts and inputs the Model needs from existing sources.
Relationships
How information, assumptions, dependencies, and evidence connect.
Logic
The calculations, rules, and methods that make the Model computational.
02
Built for AI
Proxa works like a harness, turning an LLM from a one-off chat into a system where your team and AI can work on the Model together over time. It provides the context, tools, state, permissions, and workflows needed for AI to do real work — guided, reviewed, and validated by the team.
Dive deeper →
Agents
Perform specific tasks using the Model, tools, and permissions available to them.
Loops
Repeatable workflows that monitor, validate, update, and act on the Model over time.
Tools
Skills, plugins, calculations, and actions agents can use to work with the Model.
03
Collaborative Workspace
The Workspace gives the team one persistent place to work with the Model — asking questions, running analysis and scenarios, reviewing changes, and managing outputs. It operates within a sandboxed environment with defined boundaries around data, tools, permissions, and actions.
Dive deeper →
Collaboration
Different team members can work on their part of the Model while staying connected to the whole.
Governance
Permissions, ownership, history, and review keep changes controlled and the Model trustworthy.
Security
AI operates within defined boundaries, with access only to the data, tools, and actions required for the work.
FAQ