Secure AI Agent · Data Center
AI applied to
the SLA you signed
A specialist configured to measure availability rigorously, calculate credits exactly as the contract requires, keep critical maintenance on the calendar, and warn when capacity closes in on demand.
Companies that trust UPX
Capabilities
What the Data Center Agent
can do
Main areas of work for the Data Center and Colocation AI Agent in your operation.
SLA measurement
Computes availability per client with window and evidence.
Credit calculation
Applies the contract rule and shows the math.
Critical maintenance
Tracks generator, UPS, and cooling maintenance plans.
Capacity
Projects power, rack space, and cooling against demand.
Incident records
Builds the event timeline for the client report.
Access control
Organizes requests, authorizations, and visit logs.
Skills
Capabilities that compose the specialist
Skills add specific capabilities to the Secure AI Agent according to the processes it needs to execute.
- Availability computation
- Calculates the metric over the contracted window, with evidence.
- Credit rules
- Applies the contract's table to the downtime.
- Capacity projection
- Cross-checks consumption and growth and flags the limit.
How it worksFrom the recorded event
From the recorded event
to the calculated credit
Connect your sources, define what the agent can do, and close the SLA month with the math on the table.
- Connect monitoring and contracts
- The agent works with the measurement that already exists.Connect the monitoring, the incident history, the SLA contracts, and the capacity controls so the Secure AI Agent has access only to what it needs.
- Ask for the work
- Talk to the agent in natural language.Ask for the month's SLA per client, the credit owed for an event, or the capacity projection through the available channels. The agent applies the contract rule and shows the math.
- The agent measures. Operations execute.
- From measurement to intervention, with control.The agent computes, calculates, and documents within the defined permissions. Opening critical windows, intervening in power or cooling, and granting physical access stay with the team and their own procedure.
01
02
03
Integrations
Connected to measurement and alerting
The Data Center Agent can query your metrics and run tasks in the systems your team already uses.
Grafana Prometheus Datadog PagerDuty Statuspage
Flow
What goes in,
what the agent does, and what comes out
From the downtime event to a defensible report, following the contract and the operation's permissions.
Inputs
- Event historyLOG
- SLA contractPDF
- Consumption and capacityCSV
Processing
Secure AI Agent
Processing the task
- Read
- Compute
- Calculaterunning
- Document
Contract rule appliedEvidence cited
Output
Completed
Computation prepared
- Availability per client with evidence
- Credit calculated by the contract rule
- Capacity limit flagged
Control
You define how far the Agent can act
Different actions can operate with different autonomy levels, always within the operation's procedures.
- 1
Query
Reads events and contracts and answers with the evidence.
- 2
Prepare
Computes the SLA and calculates the credit owed.
- 3
Request review
Waits for approval before notifying the client.
- 4
Execute
Performs the action within the defined limits.
Levels are configured per type of action. Opening critical maintenance windows, intervening in power or cooling, and granting physical site access always stay with the responsible team, under their own procedure.
Get started
Put a Secure AI Agent to work.
Start on the platform or choose the plan that fits the pace of your operation.
Security and compliance
Security that can be verified.
Certifications and attestations
UPX maintains SOC 2 Type II and ISO 27001, with independent audit over its information security controls.
Privacy and regulation
- LGPD
- Operations follow Brazil's Law 13.709/2018. In AI Agent contracts, UPX acts as data processor; the legal basis remains with your company.
- Zero Data Retention
- A product policy, not a certification: with compatible providers and configurations, processed content is not retained after execution.
- Retention and deletion
- The retention policy is defined by contract. Once the contract ends, data is deleted within the agreed period.
Frequently asked questions
Common questions about the Data Center Agent
What operators usually ask before putting an agent in charge of SLA computation.
Can the credit calculation come out against us?
Yes, and that is the point. The agent applies the contract rule to the real measurement and shows the math. A number that hides downtime serves no one: when a credit is owed, the report says so with the evidence.Does it open critical maintenance windows?
No. It keeps the plan on the calendar, warns what is coming due, and prepares the communication. Opening windows and intervening in power or cooling follow the operation's procedure, with human approval.Does it grant physical site access?
No. It organizes the request, verifies authorization, and keeps the visit log. Clearing a person into the site is physical control and is not delegated to an agent.How does it measure availability?
Over each client's contracted window, using the measurement source you connected, with the event and timestamp cited. When two sources disagree, it shows the discrepancy instead of picking the convenient one.Does it help with capacity planning?
Yes. It cross-checks power consumption, rack occupancy, and thermal load against observed growth and warns when demand closes in on the limit — early enough for the investment decision.Is client data used to train models?
No. Content processed by Secure AI Agents is not used to train UPX models or third-party models.Can we audit what the agent did?
Yes. Every action is logged: what was queried, which rule was applied, when, in which system, and under which permission. The history stays available for review and auditing.
Secure AI Agent
Bring a Secure AI Agent to your data center
Talk to our specialists and see how to adapt this AI Agent to your processes, systems, and needs.