Enterprise platform for AI employees
AI employees for enterprise teams.
Create, run and train AI employees that work inside your processes and your systems.
- 09:14 Ticket assigned to Atlas Named
- 09:18 Plan ready for review Needs person
- 09:22 Approved by a named reviewer Approved
- 09:23 Sandbox created Policy clear
- 11:40 Draft PR opened on polis/deps-update Your review
Why Polis
- 01
Full integration
AI employees work in your existing systems and processes.
- 02
Governed autonomy
The company sets how far the AI may go on its own.
- 03
Enterprise security
Isolated installations, deterministic permissions, protected secrets.
- 04
Growing knowledge
AI employees use company knowledge, specialise and improve under human control.
01What Polis is
AI employees that work in your processes and your systems, next to your people.
A full member of your team: its own accounts, its own tasks, and work that is reviewed like anyone else’s.
- 01
Takes tasks from the systems your team already uses
- 02
Reads the requirements and asks for what is missing
- 03
Talks to people in your existing channels
- 04
Plans the work and carries it out
- 05
Hands the result to a person for review
- 06
Works through review comments and finishes the job
02How Polis works
An AI employee takes a task from Jira to a reviewed pull request, in the tools your team already uses.
No new process for AI. It joins the one you have.
- 01 · Jira
Ticket assigned
To the AI employee’s own account.
- 02 · Slack · L1
A person approves the plan
It posts a plan. A named person approves it before any code is written.
- 03 · Sandbox
Code and tests, isolated
No integration credentials or provider keys, and outbound traffic only to an allow-list.
- 04 · GitHub
Draft PR on polis/*
With a test summary. Polis never merges.
- 05 · Audit
Your developers review
As usual. Every step lands in the audit log.
03Three levels of autonomy
The company decides how independently each AI employee may work.
The company sets autonomy, never the AI. Each project has a level, each AI employee a cap, and the lower of the two applies.
-
L0 · Advisory
Analyses the work and advises. Changes no code.
-
L1 · Supervised
A named person approves the plan before any code is written.
-
L2 · Autonomous
Implements on its own. A person still reviews the pull request.
No AI employee runs above its project’s level, each AI employee can be capped lower, and approvals can be required for specific actions across the organisation.
04AI employees
One job per AI employee, and only the rights that job needs.
Each AI employee has its own named account, roles, limits and audit trail. The same rules apply to every role.
- 01 · Engineering
AI Developer
Takes a Jira ticket assigned to its own account and posts a plan. A named person approves the plan before any code is written.
- Isolated sandbox
- Draft pull requests only
- Short-lived credentials
- Cost cap per task
- 02 · Access management
Atlassian access manager
Handles access requests in Jira and Confluence: groups, product access and Confluence space permissions. Changes are proposed as a list, approved by a person, then applied and verified.
- Changes shown as data
- Approval in the panel or Slack
- Test mode, verify and undo
- Verified identity
- 03 · Next
After the pilot: more roles, under the same rules.
- GitHub access manager
- More roles
05Built for the enterprise
Every customer gets a dedicated installation, hosted where its infrastructure rules require.
Polis adapts to your infrastructure. Your company does not have to adapt to the platform.
- Full isolation. Data, configuration and integrations are never shared between customers.
- Adaptable. The installation can be fitted to your security requirements.
- Your control. You keep control of your infrastructure and your data.
06Security by code
Code makes every security decision. The language model only proposes.
The AI decides how to do the task. Polis decides which actions it may take.
- Check
- Policy
- Credential
- Execute
- Jira
- Slack
- GitHub
- 01 · Permissions
Policy and approvals
Policies per project and action. Where a policy asks for approval, a named person decides and the decision is logged.
- 02 · Secrets
Short-lived credentials
Short-lived tokens issued per call, limited to one repository for GitHub. Jira and Slack use the connection’s stored credential, which is never exposed to the AI.
- 03 · Independent of the AI
Never, by code
Enforced in code, and no setting changes it: an AI employee never merges to the default branch, never deploys, and never changes Polis’s own settings.
07Three levels of knowledge
The longer an AI employee works with you, the better it knows your processes and standards.
- 03 · Employee knowledge Its specialisation, its instructions, its experience
- 02 · Project knowledge Architecture, technical docs, requirements, conventions
- 01 · Company knowledge Standards, processes, policies, shared documentation and terms
Onboarding
When an AI employee is created, an administrator gives it the documentation and knowledge it needs, so it starts prepared for your company and the project.
Learning from experience
The AI proposes lessons from finished tasks, feedback and decisions, and a person approves each one before it is used. You can edit, retire or delete any lesson.
Book a demo
See it on your own workflow.
A 30-minute call: about 15 minutes of live product, including the pause-all switch, then your questions. Bring your CISO if you like.
What we show
- Roles that can’t overreach, with step-up for sensitive actions
- A model provider locked to a data class
- Approvals by a named person, in Slack or the panel
The pause-all switch
One switch pauses all AI work for everyone, and a banner shows it.