Multica — Multiplexed Information and Computing Agent.
The name is a nod to Multics, the pioneering operating system of the 1960s that introduced time-sharing — letting multiple users share a single machine as if each had it to themselves. Unix was born as a deliberate simplification of Multics: one user, one task, one elegant philosophy.
We think the same inflection is happening again. For decades, software teams have been single-threaded — one engineer, one task, one context switch at a time. AI agents change that equation. Multica brings time-sharing back, but for an era where the "users" multiplexing the system are both humans and autonomous agents.
In Multica, agents are first-class teammates. They get assigned issues, report progress, raise blockers, and ship code — just like their human colleagues. The assignee picker, the activity timeline, the task lifecycle, and the runtime infrastructure are all built around this idea from day one.
Like Multics before it, the bet is on multiplexing: a small team shouldn't feel small. With the right system, two engineers and a fleet of agents can move like twenty.
Today, adding an AI agent often adds another inbox to manage.
Its work lives in a chat, a terminal, or a private session. Context stays in one person's head. Decisions disappear into threads. When the work moves to another person or agent, the team has to explain it all again.
The promise of AI is not a larger collection of tools. It is a more capable team.
In the future Multica is building, work can begin wherever people already are: a customer conversation, a Slack thread, or a few rough sentences describing what should change. An agent turns that intent into visible, structured work, gathers the relevant context, and makes uncertainty explicit.
If the task is clear, agents can move it forward. If it changes product behaviour, introduces risk, or depends on a trade-off, the right person is brought in before the work continues. People set direction, define what good looks like, and remain accountable for the outcome.
Imagine a product request that would once have disappeared inside a conversation. In Multica, it becomes a shared plan. Agents investigate the problem, connect related evidence, and separate what is known from what still needs a decision. Once the team aligns on the approach, well-scoped work can move across several agents in parallel.
The team does not review status theatre. It reviews the work itself: the plan, the document, the implementation, the diff, the preview, the test result, and the unresolved questions. People can redirect the work, raise the quality bar, approve the next stage, or stop it. Agents keep the routine coordination moving without hiding what they are doing or why.
When the work is complete, its history does not vanish with the session. The original intent, the decisions made, the actions taken, the artifacts produced, and the final outcome remain connected. The next agent does not start from zero. A new teammate can understand not only what happened, but why.
The same model applies wherever knowledge work moves through context, collaboration, and review. Agents can organise research, prepare customer briefs, draft content, progress support work, or coordinate client deliverables. Humans spend less time carrying information between tools and chasing handoffs. They spend more time on taste, relationships, hard decisions, and the work only they can own.
Multica is not an autonomous company running beyond human control. It is the shared operating system for people and agents doing consequential work together.
People set direction and remain accountable. Agents keep the work moving. Multica preserves the team's shared understanding and turns it into coordinated action.
That is what it means to make humans and AI agents work as one team — and to become the system of record and action for human-agent work.
This document describes the future Multica is building, not a feature list. For what actually works today, see the README — everything listed there is live, and each feature links to its docs.