How Our Agentic SOC Works: Full Lifecycle SecOps Platform
Skip to content

The Full Lifecycle AI SOC

ThreatDefence’s full lifecycle AI SOC brings security data, detection engineering, AI investigations, human-approved response and verified improvements together on one SecOps platform. You get continuous security operations backed by our 24x7 SOC, with evidence and decisions carried from the first signal through to resolution.

Analyst hub in the ThreatDefence SecOps platform for acme-corp: alerts, investigations, cases and data source health in one view

One Lifecycle, One Platform

ThreatDefence runs the full security operations lifecycle on its own platform. Data collection, detection, investigation and remediation work as connected stages, so the outcome of one becomes the input for the next.

Each stage has a distinct job:

Data foundation:

Our own security data platform collects, normalizes and monitors telemetry across your environment, without relying on a third-party data lake.

Detection and readiness:

Detection engineers maintain our in-house ruleset, while AI adapts detection content to the activity and conditions in each environment.

Investigation and understanding:

AI triages every alert, correlates evidence across sources and records an explained verdict for analyst review.

Remediation and improvement:

Findings requiring action become owned cases, with fixes tracked and improvements verified against the controls that identified the gaps.

The lifecycle continues after an incident closes. Investigation outcomes inform detection tuning, while completed remediation changes the environment being monitored.

// last 30 days across the ThreatDefence platform
20BEvents ingested
4.8MAlerts processed
25,000+AI verdicts written, each one explained
1,100+Detection rules in production
14,000+Systems reporting in

Build the Data Foundation

ThreatDefence collects security telemetry from your existing tools into a data platform that we built and operate ourselves. Endpoint, network, firewall, identity, email, cloud, vulnerability and attack surface data, together with security alerts, become available through one backbone.

ThreatDefenceFAIR 62Search this page…secops › acmeacme-corp24h
Data Sources
Every feed watched for silence, volume change and new data
38Sources connected
36Reporting normally
1Volume drop
1Silent
2New feeds found
SourceTypeVolumeLast eventStatus
Microsoft 365 auditidentity2.1M / day1 min agoHealthy
Microsoft Defenderendpoint412K / day2 min agoHealthy
FortiGate syslogfirewall6.8M / day38 min agoVolume down 61%
Okta system logidentity0 / day4 h 12 m agoSilent, gap measured
NDR sensor, Sydney DCnetwork3.4M / day1 min agoRecovered 09:14
AWS CloudTrailcloud188K / day3 min agoNew feed, dashboard built
Okta system log stopped at 05:02. The gap is recorded against detection coverage for identity, the owner was notified, and recovery will be confirmed when events resume.
One security data backbone

Telemetry from different vendors is parsed and normalized in our own platform, without a third-party data lake underneath or unexpected per-query charges.

Guided source onboarding

Each source connects through a guided process. When a tenant starts sending a new feed, the platform automatically generates dashboards named according to the data they contain.

Continuous source monitoring

Every tenant’s feeds are monitored for collection health. Silent or interrupted sources are flagged, the missing-data period is measured, and recovery is confirmed.

A connected source and a healthy source are different things. The platform monitors whether evidence continues arriving, so collection problems can be identified rather than allowing detection coverage to deteriorate unnoticed.

Keep Detection Content Current

ThreatDefence develops and maintains its own detection content, combining detection engineers with decades of SOC experience and AI-assisted tuning. Rules are maintained against observed adversary behavior and adapted to the activity in each environment.

ThreatDefenceFAIR 62Search this page…secops › acmeacme-corp24h
Detection Library
In-house rules mapped to MITRE ATT&CK, tuned against what acme-corp actually sends
1,142Rules in production
14ATT&CK tactics covered
23Tuned this week
4Proposals for review
9Kept as customer tuned
RuleTechniqueSourceState
Inbox rule forwarding to external domainT1114.003emailUpdated, global library
Impossible travel sign-inT1078.004identityTuning proposed: VPN egress
LSASS memory access by unsigned processT1003.001endpointHealthy
OAuth app granted mail read to all usersT1528cloudNew rule, testing
Beaconing to newly registered domainT1071.001networkCustomer tuned, kept
Shadow copies deletedT1490endpointHealthy
Tuning proposal: exclude acme-corp VPN egress from impossible travel. Built from 14 triage verdicts this week, every one confirmed benign. Waiting for a detection engineer to approve.
In-house detection engineering

Our team builds and maintains detection rules across identity, endpoint, network, email and cloud. Content is mapped to MITRE ATT&CK tactics and techniques and updated as adversary behavior changes.

Environment-specific tuning

AI uses the telemetry and investigation outcomes from each environment to tune and recategorize detection content. Proposals derived from triage outcomes are reviewed by a detection engineer before application.

Continuous library improvements

Updates to our global detection library are distributed to installations as product updates, rather than relying on occasional content releases.

Preserved customer changes

The platform records where each tuning change originated. Rules adjusted by your team are preserved rather than overwritten by subsequent updates.

Tuning From Real Verdicts

Each proposal comes from triage outcomes on your own alerts. A detection engineer reviews it before it applies.

AI tuning suggestions in the ThreatDefence portal: rules ranked by noise, each with its trend, firing count and proposed change

Detection improvement is connected to investigation. When alerts are confirmed as benign, their recorded outcomes can inform tuning proposals, while the change history remains visible.

Measure Security Readiness

ThreatDefence measures security readiness using live protection coverage, identified weaknesses and supporting evidence. The Security Score shows how the result is calculated, with gaps affecting the score rather than receiving a passing status by default.

Risk Overview in the ThreatDefence portal for acme-corp: the Security Score against its target, how the score is worked out and the fixes that return the most points

The platform also assesses all 153 CIS Controls v8 safeguards, Essential Eight maturity levels for Australian organizations, and NIST CSF 2.0 at full control depth across functions from Govern through Recover. Assessments remain connected to the underlying evidence and changes in the environment.

Explore cyber risk management >

Investigate Every Security Alert

ThreatDefence’s AI analyst triages every alert, correlates and deduplicates related signals across connected sources, and records its verdict with supporting reasoning. The process runs around the clock rather than depending on an analyst manually selecting alerts from a queue.

ThreatDefenceFAIR 62↓7.0Search pages…secops › acmeacme30d61
Agent runlive
alertImpossible travel sign-in, finance user
1Entra ID sign-insj.nguyen signed in from Sydney, then from Lagos 41 minutes later0.9 s
2Identity historyNo travel booked, no earlier sign-in from that country in 18 months1.2 s
3Network contextLagos address is a hosting provider, outside acme-corp VPN egress0.8 s
4Entra ID tokensSession token reused from a device that was never registered1.6 s
5Microsoft 365 auditMailbox searched for “invoice” and “bank” two minutes after sign-in2.4 s
6Endpoint (EDR)No suspicious activity on the user’s laptop2.1 s
7Case historyNo open ticket or approved exception for this user0.5 s
VerdictTrue positive
SeverityHigh
Confidence0.92
ATT&CKT1078.004
Time36 s
1Revoke all sessionsheld for approval
2Open a case with the evidence attachedheld for approval
3Confirm with an analyst before closingheld for approval

The AI assessment and human decision have separate roles. An AI verdict provides the investigation outcome and supporting evidence, while a SOC analyst confirms any compromise before it is treated as established.

Actions affecting systems outside the platform require human approval. The analyst can review the evidence, confirm the findings and authorize the next step before a response is carried out.

Explore AI alert triage >

Run Guided Security Investigations

ThreatDefence provides guided investigations that take an analyst from an alert through a repeatable set of evidence checks. Expert-written investigation recipes cover business email compromise, stolen session tokens, persistence, lateral movement and data exfiltration.

Run the Investigation Checks

Each recipe executes its checks against your own telemetry using read-only queries. Launching an investigation from an alert carries the original alert context into the process, avoiding the need to reconstruct it manually.

One Click From Any Alert

An investigation launched from an alert carries the full alert context with it.

A guided hunt in the ThreatDefence portal: each recipe check with its result, the highest severity findings and the queries behind them

Review the Investigation Timeline

The investigation presents the checks performed alongside the evidence found. Analysts can inspect the results, see which findings carry the highest severity and follow the sequence of activity through the resulting attack timeline.

Timelines and Blast Radius

The investigation builds the attack path and timeline, and overlays what the hunt checked against what it found.

An attack timeline in the ThreatDefence portal, with each step of the attack and the checks the hunt ran placed on the same picture

Share the Investigation Findings

Investigation results can be exported as branded reports. The report presents the findings in a format customers can understand, while the underlying investigation retains the queries and checks used to reach those conclusions.

Guided investigations turn expert knowledge into a repeatable process without removing the analyst’s ability to inspect the evidence.

Turn Findings Into Fixes

ThreatDefence connects investigation findings and remediation work through case management. Issues requiring action are assigned to a person, tracked through resolution and linked to the evidence that established the need for a fix.

The case board in the ThreatDefence portal for acme-corp: new, investigating, waiting for approval, escalated and closed cases with severity, owner and status
Cases with evidence and ownership

Escalations, review actions and risk register entries automatically open cases with supporting evidence and an assigned owner.

Visible investigation progress

Cases show their current status, including new, investigating, waiting for approval, escalated and closed. The case record is updated when the associated work is completed.

Verified improvements

The platform uses CIS safeguard assessments to identify gaps and track remediation. When a fix changes the environment, the underlying assessment and Security Score reflect the improvement.

Ongoing follow-through

Review actions remain tracked between assessments, so outstanding work continues through the operational workflow rather than being left in a report.

Broader cyber risk management, including virtual CISO services, risk registers, insurance readiness and executive reporting, extends this process into a continuing security improvement program.

AI Roles Across Operations

ThreatDefence uses specialized AI personas for different parts of security operations. Each has a defined role, from checking security controls to running investigations, answering questions and following cases through to closure.

The CISO

Reviews security posture, prepares recommendations and follows up on outstanding improvements, drawing attention to actions that would have the greatest effect on the Security Score.

The Experts

Specialized network, email, identity and endpoint agents examine the data associated with their areas. They run defined control checks and identify which conditions pass and which need remediation.

The Hunter

Executes expert-written investigation recipes, completes the associated checks and presents the evidence, attack timeline and findings for analyst review.

The Chat

Answers questions about the data displayed in your current portal view. It runs read-only queries against supporting evidence and returns the results within your existing access permissions.

The Backend

Tracks operational work from the initial event through to closure. It monitors escalations, checks whether they have been acknowledged and identifies work that has stalled between stages or shifts.

Ask Anything About Your Data

The chat queries the evidence read-only and answers with the numbers it found.

The AI assistant in the ThreatDefence portal answering a question about MFA push fatigue for acme-corp with the queries it ran

These roles operate within the same platform and use its security data and workflows. Analysts retain responsibility for confirming compromises and approving actions that affect external systems.

From Event to Verified Fix

The full lifecycle AI SOC connects the initial event with the investigation, response and subsequent improvement. The following fictional example shows how the stages work together for acme-corp.

1
The data arrives:A finance user’s sign-in reaches ThreatDefence through the identity feed. Source monitoring shows that the feed is healthy and reporting normally.
2
A detection fires:An impossible travel rule identifies sign-ins from Sydney and Lagos, 41 minutes apart. The second address is outside every network used by acme-corp, prompting further investigation.
3
The AI analyst investigates:In this example, the AI spends 36 seconds checking the user’s sign-in history, session token, mailbox and laptop. It records a true positive verdict with the evidence and reasoning behind that assessment.
4
A person confirms:A SOC analyst reviews the findings, confirms the compromise and approves revocation of the affected user’s sessions.
5
The fix gets an owner:A case opens with the evidence attached and an owner assigned at acme-corp. The work includes resetting the password and correcting the sign-in policy that allowed the session.
6
The improvement is verified:Once the policy is corrected, the Security Score reflects the recovered points. The investigation verdict also becomes an input for subsequent detection tuning.

The original detection, supporting investigation, human decision and remediation outcome remain connected. The same case provides a record of what happened, what was authorized and what was changed.

See the Whole Lifecycle in One Session

Bring one of your noisiest alert sources. We will show you the path from raw telemetry to a verified fix, on the platform, end to end.

What the session shows
  • Raw telemetry from your noisiest source
  • The detections and AI verdicts it produces
  • A guided investigation with its timeline
  • A finding taken through to a verified fix

Cover Your Entire Environment

ThreatDefence provides coverage across endpoints, cloud systems, networks, remote users and the security tools already in your environment. Existing integrations supply telemetry, while our own agents and sensors provide deeper visibility where required.

Deployment takes hours for most environments. Detection content and threat intelligence are included, and the service uses a simple pricing model.

Forensic Collection in the ThreatDefence portal for acme-corp: sources selected, files collected, findings from the AI analysis, evidence by source, collection progress and a fleet hunt
Endpoint threat detection and DFIR

The ThreatDefence agent deploys in a few clicks and provides endpoint telemetry for detection, live response and digital forensics.

Cloud monitoring and assurance

Microsoft 365, Azure, AWS and Google Cloud are monitored for threats, vulnerabilities and misconfigurations, with ongoing cloud assurance capabilities.

Network detection and response

NDR sensors analyze network traffic to identify lateral movement and produce high-fidelity network security alerts.

Hacker deception

Honeypots and honeytokens provide evidence of attacker interaction, with each action recorded for investigation.

The platform connects these different sources through the same detection and investigation workflow. Analysts can examine activity across the environment rather than treating every security product as a separate investigation.

Explore endpoint visibility and DFIR >

Preserve Evidence Across Sources

ThreatDefence records security evidence across endpoint, network and cloud environments for real-time inspection at scale. When preventive controls are bypassed, those records support investigation, containment and forensic analysis.

Different sources contribute different parts of an investigation. The value of a unified security operations platform comes from examining those records together.

Evidence sourceWhat it contributes to security operations
Existing EDR and ThreatDefence agentEndpoint detections, additional host visibility, live response and forensic collection.
Identity, email and cloudActivity that can be correlated across different stages of an attack.
Network sensorsNetwork traffic visibility, lateral movement findings and supporting evidence for network hunting.
Honeypots and honeytokensRecorded attacker interaction that can be investigated alongside other security signals.
Guided threat huntingRepeatable checks across endpoint, cloud and network telemetry, with findings and investigation timelines.
Case managementEvidence, assigned ownership, investigation status, approvals and the work required to resolve findings.

The same evidence can support several activities. An endpoint record may contribute to a detection, become part of a guided investigation and later support the forensic analysis of a confirmed incident.

Keeping that context together helps the SOC explain its conclusions and follow the response through to completion.

Everything in One Platform

ThreatDefence delivers its security operations capabilities through one managed SaaS platform. The service combines security technology, AI-assisted workflows and a 24x7 SOC.

Detection and analytics

SIEM, XDR, NDR, threat intelligence and automation, including MITRE ATT&CK mapping, daily detection updates, machine learning and user behavior analytics.

Response and forensics

Investigation and response playbooks, automation, digital forensics tools, threat hunting artifacts and 24x7 incident response.

Exposure and risk

Vulnerability management, attack surface management, curated threat intelligence, dark web monitoring, cloud monitoring and assurance, asset inventory and configuration benchmarking.

Security operations

Alert management, escalation workflows, SIEM and 24x7 SOC services, hacker deception and ongoing platform management.

Reporting and data

Real-time dashboards, a customer portal with live reports, customizable and compliance reports, log management and data retention.

The capabilities share a common operational foundation. ThreatDefence maintains the platform and provides the SOC expertise behind it, while your team retains visibility into findings, investigations and the actions requiring its involvement.

Explore the AI SOC platform >

Talk to Our Experts

See how ThreatDefence automates security monitoring, compliance and cyber risk management in one platform.

    I'm an MSP, MSSP, or reseller

    By submitting this form, you accept the terms in our privacy policy.

    Frequently Asked Questions

    What is a full lifecycle AI SOC?

    A full lifecycle AI SOC manages the connected stages of security operations, from collecting security telemetry and maintaining detection coverage to investigating alerts, coordinating response and verifying improvements. ThreatDefence runs these stages on its own platform, with AI handling repetitive investigation work and a 24x7 SOC providing human oversight. Investigation results also inform future detection tuning, so the process continues beyond individual incidents.

    What is a SecOps platform?

    A SecOps platform combines the tools and workflows used to collect security data, detect threats, investigate suspicious activity, coordinate response and report outcomes. Rather than requiring analysts to move between separate systems for each stage, a unified platform keeps related evidence and operational work together. ThreatDefence extends that model with its own security data platform, AI agents, detection engineering and managed SOC services.

    Does it work without agents on endpoints?

    Yes. ThreatDefence can collect security data from supported services through APIs and log forwarding, including Microsoft 365, cloud platforms, identity providers, firewalls and existing EDR products. The ThreatDefence endpoint agent is optional. Installing it adds deeper host-level visibility, live response and forensic collection, so the decision depends on the level of endpoint evidence and response capability required.

    What does the AI decide on its own?

    ThreatDefence’s AI triages every alert, correlates related signals and records a verdict with supporting reasoning. It also supports detection tuning, guided investigations and case creation. These assessments remain subject to human oversight. A SOC analyst must confirm a compromise, and containment or another action affecting systems outside the platform requires human approval under your policy. AI can prepare the work without taking those decisions away from people.

    How long does deployment take?

    Deployment takes hours for most environments, with guided onboarding for supported data sources through APIs and log forwarding. Detection content and threat intelligence are included, and dashboards are generated as new feeds begin sending data. Monitoring coverage starts on the first day and expands as additional sources are connected. Collection health monitoring helps identify sources that are interrupted or have stopped reporting.

    Where is our data stored?

    ThreatDefence stores your data in the region you choose, Australia or the United States. Data remains in the selected region and does not move between the two. Australian customers have data sovereignty guaranteed in Australia, with their security alerts handled by ThreatDefence SOC analysts located in Australia. The regional choice determines where customer data is stored and processed within the service.