FAQ

Getting Started and Infrastructure

What is the infrastructure cost of running the DuploCloud AI Suite?

Infrastructure costs (EC2 instances, load balancers, S3 buckets, Bedrock LLM cost) will be approximately $30-50 per month. LLM/Bedrock costs will only apply when you actively use the features and will be a few dollars per month.

Available Agents and Customization

What agents are available out-of-the-box?

We provide several pre-built agents including Kubernetes troubleshooting, Private GPT, Architecture Diagram generation, and Observability. Full documentation is available in the out-of-the-box agents section.

Can we get custom agents built for our specific use cases?

Yes, we can develop custom agents tailored to your specific DevOps use cases.

What is Private GPT and why should we use it?

Private GPT is a ChatGPT-like interface that keeps all your data within your AWS account. It's ideal for organizations with data security requirements that prevent using public AI services.

AI Models and Framework

Does DuploCloud use Artificial Intelligence (AI) models?

Yes. DuploCloud AI Suite is a framework to build and deploy AI agents. These agents will use LLMs to respond to users, take action, etc.

What Model Providers do you support?

The default Model Provider will be AWS Bedrock to keep the data within your cloud account. With our PreBuilt Agent option any Model Provider can be used.

What is the purpose of these AI models?

The models, provided through AWS Bedrock inside your own AWS account, are used to power your custom AI agents. They generate intelligent responses to support DevOps workflows, troubleshooting, and automation.

What type of output do the models generate?

Outputs include infrastructure provisioning commands, log analysis, troubleshooting recommendations, and other data that help DevOps teams operate more efficiently.

Data and Training

What data is used to train the AI models?

We do not train models. DuploCloud uses foundational models from AWS Bedrock. Your data is never used to train or improve these models.

What data is input into the models?

The platform provides context about your infrastructure (e.g., Kubernetes cluster names, namespaces, application logs) so the model can assist with relevant operations.

Does DuploCloud's AI process personal information?

No. The AI agents do not process personal or sensitive personal information.

How does DuploCloud ensure the security of my data in DuploCloud AI Suite?

The DuploCloud Platform has always been a self-hosted solution that lives inside your Cloud - The DuploCloud AI Suite follows the same pattern. The AI Agent, any knowledge sources (VectorDBs), and the LLM are all hosted inside your cloud account.

System Integration and Oversight

What systems do AI agents connect to?

DuploCloud's AI agents interact with your DevOps and infrastructure systems (such as AWS, Datadog, and application logs) to perform tasks and provide insights.

Is there human oversight of AI actions?

Yes. The DuploCloud AI HelpDesk is designed with human-in-the-loop oversight. Users can review, approve, or reject actions the AI proposes before execution, ensuring control and safety.

Business Value

Why should we try the Agentic Help Desk?

The Agentic Help Desk tackles the biggest DevOps headaches with AI-driven workflows:

• Ticket overload → Automates repetitive tasks so engineers focus on higher-value work

• Troubleshooting bottlenecks → Provides end-to-end visibility to accelerate root cause analysis

• Deployment delays → Manages rollbacks, approvals, and scaling with human-in-the-loop safeguards

• Knowledge silos → Captures and retains critical expertise, ensuring continuity when people leave

With agents for Kubernetes, observability, and log analysis, teams already see faster fixes, safer launches, and less burnout.

Future Development

What's coming next? What more can I expect?

We're expanding the HelpDesk with more prebuilt agents to tackle problems across your infrastructure, and new capabilities designed to deliver long-term value for your growth and modernization journey.