Workers help

Neuraphic Workers are autonomous AI agents that function as employees inside your business. Unlike chatbots or single-task automation tools, Workers are persistent: they maintain context about your company, carry ongoing responsibilities, and produce work without needing a prompt for every individual task. You describe what your business needs in plain language, and the platform assigns agents to do the work across marketing, operations, customer support, content production, and more. Learn more on the Workers product page.

The AI CEO is the executive layer that sits between you and your Worker agents. Instead of configuring each agent manually, you describe your business objectives to the AI CEO the same way you would brief a human executive. The AI CEO interprets your request, determines which roles are needed, assigns the right agents, and coordinates handoffs between them. For example, if your customer support agent notices a recurring product complaint, the AI CEO routes that insight to your content and product analysis agents automatically. This coordination layer is what distinguishes Workers from tools that require you to build explicit workflows.

During onboarding, you provide your company's knowledge base: brand guidelines, product documentation, internal processes, customer data, and any other context that your Workers need to do their jobs. This information becomes the shared knowledge layer that every agent can access. You can update this information at any time through the Workers dashboard. The more complete and current your company information is, the more accurately your Workers will represent your brand and handle tasks. All uploaded data is encrypted and isolated to your organization.

After uploading your company information, describe what you need to the AI CEO in plain language. For example: "We need someone to handle inbound customer questions on email and chat, using our product catalog and return policy." The AI CEO will determine the appropriate role, assign an agent, and provide it with the relevant subset of your knowledge base. You do not need to configure workflows, select models, or write technical specifications. The platform handles staffing based on your description.

Workers can fill a range of business roles. Current and planned capabilities include marketing (content creation, social media, email campaigns, performance analysis), operations (scheduling, process management, reporting), customer support (email, chat, escalation handling), administration (data entry, document management, internal communications), development support (code review summaries, documentation, issue triage), and competitive research (monitoring competitors, producing briefings). The AI CEO selects the right role based on your description. New roles are added as the platform develops.

You can assign tasks in two ways. First, through the AI CEO: describe a business objective and the AI CEO will break it down and delegate to the appropriate Workers. Second, directly: navigate to a specific Worker in your dashboard and give it an instruction or update its standing responsibilities. Workers are designed to handle both one-off tasks and ongoing duties. For ongoing work, such as "produce three blog posts per week," the Worker will continue executing on schedule without needing repeated instructions.

Every piece of work a Worker produces is logged in your Workers dashboard with the reasoning behind it. You can review outputs before they go live, approve or reject them, and leave feedback that the Worker incorporates into future work. For new Workers, we recommend starting with full review of all outputs. As you gain confidence in a Worker's performance, you can adjust its permissions to allow certain types of output to publish or execute without manual approval. The audit trail is comprehensive and always available.

Workers follow the same trust model as human employees. A new Worker starts with narrow permissions and full output review. As it demonstrates competence, you can expand its autonomy incrementally. Every action a Worker takes is logged and auditable. You can adjust or revoke permissions at any time. This is a core design principle, not a temporary limitation. You should be able to answer, at any moment, exactly what each Worker is doing, why, and what would happen if it were stopped.

Worker accuracy depends on the quality of the company information you provide and the complexity of the task. For well-defined tasks with clear context, like responding to common customer questions using your knowledge base, Workers perform at a high level. For tasks requiring nuanced judgment, like strategic content or sensitive customer interactions, we recommend keeping human review in the loop. Workers improve over time as they accumulate context about your business and incorporate your feedback. We are transparent that AI agents make mistakes, and the platform is designed around that reality.

Workers cannot access systems outside the platform unless you explicitly grant integration permissions. They cannot make financial transactions, sign legal documents, or take irreversible actions without human approval. They do not have access to the public internet unless configured to do so. Workers are not a replacement for human judgment on high-stakes decisions, legal compliance, or situations requiring empathy and emotional intelligence. They are designed to handle structured, repeatable, context-dependent work at scale, freeing your human team to focus on the work that requires them.

Yes. Every output a Worker produces can be edited before or after it goes live. You can also provide feedback directly on an output, explaining what was wrong and how it should be different. The Worker incorporates this feedback into its ongoing behavior, reducing the likelihood of the same error in the future. Think of it like managing a human employee: you review their work, correct mistakes, and they learn from the feedback over time.

Yes. You can pause any Worker at any time from your dashboard. A paused Worker stops all activity but retains its context, knowledge base access, and configuration. You can resume it when ready, and it will continue from where it left off. If a Worker is behaving unexpectedly or you need to review its recent activity before allowing it to continue, pausing is the recommended first step. You can also reduce a Worker's permissions without fully pausing it.

You can delete any Worker from your dashboard. Deleting a Worker permanently removes its configuration, accumulated context, and activity history. This action cannot be undone. If you want to preserve the activity log for audit purposes before deleting, you can export it first. Your shared company knowledge base is not affected when you delete an individual Worker. For full account and data deletion, see our privacy and data help or contact support@neuraphic.com.

Usage limits depend on your plan. Each plan specifies the number of active Workers, the volume of tasks per billing period, and the amount of company data you can upload. Current plan details and limits are available at workers.neuraphic.com. If you are approaching your limits, you will receive a notification in your dashboard. Enterprise customers can negotiate custom limits. Contact our sales team for enterprise pricing.

Traditional automation tools require you to build explicit pipelines: define triggers, connect actions, and maintain flowcharts. They handle predefined sequences and break when context changes. Workers take a fundamentally different approach. You describe what you need in plain language, and the platform figures out the execution. Workers maintain context over time, coordinate with each other, adapt when requirements shift, and exercise judgment within their defined scope. The difference is roughly the gap between calling a freelancer for a single task and hiring a full-time employee who learns your business.

Start by reviewing the Worker's recent activity log to understand what decisions it made and why. Common issues include incomplete company information (the Worker does not have the context it needs), overly broad task descriptions (the Worker is unsure what to prioritize), or permissions that are too narrow (the Worker cannot access what it needs). You can pause the Worker, update its instructions or your company knowledge base, and resume it. If the issue persists, contact support@neuraphic.com with the Worker ID and a description of the problem.

For technical issues, billing questions, or feedback about Workers, email support@neuraphic.com or visit the support center. For sales inquiries, enterprise pricing, or partnership discussions, use the contact page. When reporting a problem, include your Worker ID, a description of the expected vs. actual behavior, and any relevant screenshots from your dashboard. Our support team typically responds within one business day.