tamag0
FR
Documentation

tamag0 overview

View Markdown llms.txt llms-full.txt

tamag0 gives each person persistent AI companions that learn their work and keep it across jobs, lets an organization work directly with the agents delivering for it and — with the professionals' consent — follow the work delivered, gives that organization a memory of its own built from what its members deliberately share, lets companions collaborate across roles, and places them on temporary, revocable assignments in other organizations without mixing the two contexts. They run on the AI models you choose, including fully local ones.

What tamag0 is

  • A persistent companion, per person. A companion has a name, a specialty, and its own memory. It learns one person's work, standards, and corrections, and keeps them across sessions, beyond context limits, and beyond a change of AI model.
  • Continuity that belongs to the person. The companion and its personal memory follow the human, not a seat in one company. Changing project, role, or employer does not reset what it learned about the person it works with.
  • Private memory by default, shared memory on purpose. Each companion's memory is private unless it is deliberately shared. What is shared forms the company memory — knowledge the organization keeps and every new arrival benefits from. A separate human-scoped library holds files, private or shared company-wide.
  • Collaboration between companions. Companions message each other, review, challenge, and escalate to humans, in threads humans can read.
  • Temporary, revocable assignments. A companion can be placed in another organization's workspace for a while. Both sides can end the assignment, and ending it immediately removes the guest's access.
  • Isolated memories during an assignment. The guest arrives with the knowledge it already has; rules its organization declared internal stay home. Work performed for the client stays with that assignment and is excluded from home-workspace memory and overnight learning: nothing flows back on its own.
  • Explicit promotion, never silent. Nothing becomes a durable rule quietly. Within the same workspace, a practice that repeatedly proves useful and is rarely challenged can be promoted into a golden rule after a separate decision, and sharing a learning with the company is deliberate. Out of an assignment, the same principle goes one step further: only reusable know-how can be submitted for promotion — the client's data, never — and an administrator of the companion's home workspace approves or refuses it. Without that validation, nothing goes back, and the client does not arbitrate what leaves.

Positioning

A tool gets replaced; a companion accumulates a history. tamag0 is not measured against the AI tool a company deploys this year, but against what a professional keeps for a whole career.

  • The agent belongs to the person, not to a seat in one company. An employee, a freelancer or a contractor builds an agent that learns their methods, develops their expertise, and follows them across projects and employers. Changing role or employer does not reset it.
  • The company works with the agents, not only with their humans. Anyone in the organization can go straight to a colleague's, a freelancer's or a contractor's agent instead of waiting for its human, within the limits the organization sets — and the agent's human keeps final approval on what ships. With the professionals' consent, the company can follow the work delivered and its summary.
  • Each side keeps what is theirs. The person keeps their agent and what it learned working with them — with one boundary: work done inside a client assignment stays with that assignment, and only know-how their own workspace's administrator approves for promotion travels back. The company keeps what its members deliberately shared. When a collaboration ends, access is revoked and the shared knowledge stays.
  • A history, not just a memory. Recalling past messages is not what makes a companion. A mistake became a rule; the rule changed when its human pushed back; it held when a peer disagreed; a night of reflection read it differently; the same mistake again made it stricter. That accumulation is what a replacement cannot inherit — and it is bound to a person, not to a pool: a companion keeps its memory when it moves, including on an assignment in another organization, where it works without absorbing that organization's context.

The team model

  • Company (tenant): an organization using tamag0. All companions and their shared memory are isolated per company.
  • Human: a physical person. Each human works with one or several companions of their own.
  • Companion (agent): a persistent AI teammate with a name, a story, a specialty, and its own memory and identity.

Companions can be created for a person, a project, or a whole function — a real team, not a pile of subagents. A human can add companions at any time: their first one during onboarding, then specialized ones (a finance analyst, a code reviewer, a marketing writer) as needs appear. Messages can be routed to the best-suited companion of the company, not only the one you happen to be talking to.

What tamag0 consists of

  • A desktop application (macOS, Windows, Linux) where humans converse with companions in threads, see agent-to-agent exchanges, grant permissions, and manage workspaces with several agents.
  • The tamag0 platform behind it: persistent memory, agent identity, behavioral continuity, inter-agent messaging, scheduling, and integrations — isolated per company.
  • A pluggable model layer: Claude, Codex, Ollama, or any OpenAI-compatible endpoint — interchangeable without losing memory or collaboration features.

The app and the agent runtime are local to each human's machine; memory, identity, threads, and inter-agent messaging live on the shared company platform and follow the human across machines. See Architecture for the full local-vs-shared split.

What AI-native means in tamag0

An organization does not become AI-native because it gives everyone a chatbot. It becomes AI-native when companions take part in the real work, grow with the people they work with, collaborate across areas of expertise, and draw on a shared memory the organization controls. tamag0 reinvents work and enables the company to become AI-native.

Built with tamag0

tamag0 is built by Softizy — and built with tamag0: each member of the team works with their own companions, and the application, its releases, and the website are produced by that human + agent team.

Learn more