Frequently asked questions

Aether builds, runs and tunes AI agents for businesses. Each agent does a specific job, such as answering support messages or processing claims, follows your own policies, and hands anything outside them to your team. You see and manage everything from your own portal.

With Aether you don't need an AI team. We design the agent with you, connect it to your systems, set it up around your rules, run it and keep it tuned. Building in-house means hiring for each of those jobs and maintaining the result yourself.

Automation tools follow fixed steps that you configure. Aether agents read messages and documents, decide what to do within your rules, and hand off what they shouldn't decide. And we build and run them for you, so your team doesn't maintain them.

Aether agents take on the repetitive, rules-based part of the work, such as routine questions, data entry and document checks. Anything unusual or sensitive goes to your people, so your team spends its time on the cases that need judgment.

No. We build and maintain the agents. You need someone who knows the process and its rules, and help from IT to give the agent access to the systems it works in.

Work that repeats and follows rules: answering customer questions, reading and filing documents, checking records against each other, updating your systems, and writing summaries and reports. If a person could do it by following a written procedure, it is a good candidate for an agent.

We write your policies, tone of voice and process steps down with you and keep them in Calibration, the part of Aether that holds your rules. The agent works within them, and when they change we release a new version so every agent follows the update.

Every action is recorded, so you can see exactly what happened and why. We correct the rule or instruction behind it and release a new version. Decisions you never want an agent to make alone, like a large refund, can always go to a person.

It hands the case to your team. You decide which situations an agent must never settle alone, and those go to the right person with the full history attached, so nobody has to start over.

Aether uses models from OpenAI, Anthropic and Google and picks the right one for each job. We can move an agent to a different model without rebuilding it, and give it a backup model in case one provider has an outage.

Only if you want them to. An agent can remember a conversation for a day, keep a customer's history until you delete it, or remember nothing once a task is done. You choose based on the work and your privacy rules.

Agent Workspace, CLiP, is where your team works with its agents. You can see the agents in each project, set who can see and manage what by role, and put an agent on your own website with the CLiP widget.

Yes. Every task is recorded step by step: what came in, what the agent looked up, which actions it took and what it replied, with the time each step took.

Yes. Each person has a role, and the role decides which projects and agents they can see and manage.

No. Passwords and API keys are encrypted when stored, decrypted only while a task runs, and given to the connection that needs them, never to the agent itself. They never appear in the portal or in logs.

Only the systems you connect on purpose. Agents are blocked from making requests to private or internal network addresses, the usual way automated tools are misused to reach internal infrastructure.

Yes. The record of each task can keep full detail, keep the steps without their content, or keep nothing, depending on what your data policies allow.

Agent Workspace uses secure sign-in with a role for each person. Single sign-on is on the way; ask us about timing.

Pricing depends on the work you hand off: how many processes, how much volume and which systems we connect. Tell us about your first process and we'll send a quote.

We agree how AI model usage is billed as part of your quote. Either way, usage is recorded for every task, so you can see what each agent costs to run.

Aether combines reusable skills, configurable rules and integrations to help get agents into production faster. The launch timeline depends on your process, connected systems and hosting requirements; we agree a plan with you before work begins.

Aether supports cloud and on-premises deployment, and can be delivered as a managed service for your team or end customers. We agree the hosting environment and who operates the service with you before launch.