Skip to content
imper.ai

Post-Hire Verification Overview

Overview ​

Post-Hire Verification is a lightweight telemetry and risk analysis framework designed to continuously validate that corporate devices are being operated by the authorized employee and not by unauthorized third parties, remote operators or covert access infrastructure.

The platform collects and analyzes signals, generates detections and risk indicators and surfaces findings through the Devices page - including risk level and supporting evidence.

The solution ensures that:

  • Devices are continuously monitored for anomalous access patterns and operator behavior.
  • Risk signals are correlated and scored across multiple telemetry dimensions.
  • Security teams receive actionable findings without disrupting day-to-day employee workflows.

How does it work? ​

Post-Hire Verification operates as a background telemetry layer on corporate devices. Once deployed, it passively collects signals and applies risk analysis without requiring any action from the employee.

imper.ai collects these signals in one of the following ways. See Quickstart to choose the one that fits your organization.

MethodHow signals are collected
Collector scriptThe imper.ai collector script runs on managed devices through your existing MDM (Iru for Mac, Microsoft Intune for Windows), on a schedule or on demand.
Sign-in checkimper.ai checks the device while the user signs in to Microsoft Entra, before the user's own MFA.

Signal collection ​

A broad set of behavioral and environmental signals is captured, including:

  • Device activity patterns and usage cadence
  • Input behavior and interaction anomalies
  • Network access context and remote session indicators
  • Process and application telemetry
  • Environmental signals that may suggest unattended or third-party operation

Risk Analysis Engine ​

Collected signals are processed by the risk analysis engine, which applies correlation logic and scoring models to identify indicators of unauthorized access or covert operation. The engine evaluates:

  • Deviation from the employee's established behavioral baseline
  • Presence of remote access tooling or suspicious process chains
  • Temporal anomalies and session-level inconsistencies
  • Cross-signal patterns associated with unauthorized operator activity

Detection and Risk Scoring ​

When the engine identifies a meaningful risk pattern, it assigns a risk level:

Risk LevelDescription
MediumOne or more signals indicate a risk.
HighMultiple correlated indicators of unauthorized or covert access detected. Immediate action is recommended.

Risk Types ​

Risk type identifies what kind of risk indicator was detected on the device. Each risk type groups one or more detections, and a device can carry more than one risk type.

Risk typeWhat it meansDetections
Remote controlSomeone other than the employee may be operating the device remotely.Remote control hardware, such as a KVM-over-IP device plugged into the computer. Remote control software sessions, such as screen-sharing or remote-access tools. An AI agent actively controlling the device.
Suspicious softwareSoftware is set up to fake the employee's presence in calls.A virtual camera and a virtual microphone installed together, as used to run a deepfake.
Suspicious environmentThe device runs in a setting that does not fit one employee working on their own computer.The device sits on a home network with many other computers, a sign of a laptop farm. The device is a virtual machine.
Suspicious device useActivity on the device points to misuse.A known malicious repository or script on the device. A file-sharing tool used to move files off the device.
Unauthorized AccessThe device is signed in to an account that should not be on it.An account on a domain flagged as suspicious.
Emulated ActivityActivity on the device that could not have come from someone at the keyboard.User activity while the device is in a state where no local interaction should be possible, such as a keep-awake app running on a headless device.
Restricted regionThe device is in a location your organization does not allow.The device is located in a region your organization has marked as restricted.

Need help or have a question? support@imper.ai