Beyond the Alert

Darren LaCasse, Director of Threat Intelligence, Detection, & Response at Elastic, makes a case that most SOC leaders are solving alert fatigue the wrong way. Starting with critical alerts keeps teams treading water. His approach of sorting by volume first, clearing the biggest bucket, then using that momentum to ask why those alerts existed at all separates short-term queue management from the actual tuning work. He also walks through how his team built an in-house AI agent that cross-references threat intelligence against their own vendor lists, software asset inventory, and vulnerability data before it ever reaches a detection engineer, filtering hundreds of daily articles down to what is actually relevant to their environment.

Beyond tooling, Darren challenges how the industry frames the talent shortage. He does not think it is a skills problem. He thinks employers do not want to make the long-term investment in junior analysts, and that avoidance is where burnout compounds. He talks about how he leads that differently: sharing his own mistakes openly, encouraging his team to document every decision so he can back them up, and what he actually looks for when hiring (someone who has solved a real business problem creatively, not a polished resume). 

Topics Discussed:

• Reframing alert prioritization by sorting queues on volume rather than severity to build analyst momentum and reduce backlog

• Using historical alert data to identify chronic tuning problems versus one-time spikes in SOC queue volume

• Building in-house AI agents that cross-reference threat intelligence against asset inventory and vulnerability data for environment-specific relevance

• Translating threat intelligence deliverables into detection rules by running source reports through AI agents and validating against internal data lakes

• Evolving detection engineering from static, hand-built rules toward dynamic, AI-assisted scoring systems that aggregate signals into actionable investigations

• Reframing the cybersecurity talent shortage as an employer investment problem rather than a pipeline or skills gap

• Building team cultures where analysts feel safe to document decisions, admit mistakes, and take time off without guilt

• Predicting the SOC analyst role shifting toward agent management, including tuning, output validation, and QA across AI-assisted workflows

Listen to more episodes: 

Apple 

Spotify 

YouTube

ECS now operates with 12 tier-one analysts instead of 14 while triaging 30,000 monthly alerts, achieving a 40% reduction in mean time to triage for Dropzone-handled alerts. Dave Howard, Senior Director of Cyber Operations, and Jesse Mainor, SOC Manager, built a hybrid model where alert sources flow to SOAR first for initial enrichment and configured auto-closure patterns, then route remaining alerts to Dropzone for structured investigation before landing in ServiceNow with complete context. Their governance approach required SOC 2 Type 2 certification as a blocking requirement before evaluating any AI vendor to prevent downstream compliance issues. 

Dave shares how his leadership philosophy comes from his military background: servant leadership that flips the organizational pyramid upside down, empowering teams to deliver outcomes while removing roadblocks. Jesse prioritizes hiring for curiosity over credentials, looking for investigative instinct and comfort with ambiguous, incomplete data rather than training technical tools.

Topics Discussed:

• Building leadership buy-in for AI implementation by framing alert volume as unsustainable headcount scaling problem 

• Establishing SOC 2 Type 2 compliance as blocking requirement before AI vendor evaluation to prevent downstream governance failures

• SOAR-to-Dropzone architecture where SOAR handles initial enrichment before routing alerts for structured AI investigation

• Breaking linear MSSP hiring model where new clients traditionally required proportional analyst headcount to handle alert volume

• Defining POV success criteria across five operational targets: alert overload, mean time to triage, handling consistency, context enrichment, scalability

• Training separate Dropzone tenants per client environment since identical alert types require different triage logic based on context

• Reducing analyst burnout by eliminating queue-clearing pressure and enabling deep-dive investigations, threat hunting, and detection engineering upskilling

• Applying servant leadership principles from military background to flip organizational hierarchy and empower SOC teams to deliver outcomes

• Hiring for curiosity over credentials by prioritizing investigative instinct and comfort with ambiguous, incomplete data in security analysts

• Maintaining 3.2% annual attrition rate by empowering analysts, providing space for mistakes, and servant leadership approach

Listen to more episodes: 

Apple 

Spotify 

YouTube

Austin Amraen, SOC Director at CommandLink, has built SOC teams from the ground up multiple times, and his approach challenges some of the field's most accepted assumptions. He rejects the tier-one-to-tier-three analyst model entirely, arguing that the biggest capability gap in most mature stacks isn't endpoint or identity but unmonitored network traffic, and measures SOC effectiveness not by MTTD or MTTR but by whether analysts are surfacing zero-days and proposing detection methods on their own.

Austin explains why most organizations have the firewall running but nobody assigned to watch what is actually moving through it and what C2 communications, unusual outbound connections, and open ports look like when someone is finally asking "that's different, what is that?" He also gets into how he handles burnout in practice: mandatory lunch every day, one-on-ones built around career goals rather than company goals, and why process improvements that free up analyst time without reducing workload just move the problem around.

Topics Discussed:

• Rejecting the tier-one-to-tier-three SOC model in favor of hiring senior analysts who can build and adapt

• NDR as the most overlooked capability gap in organizations with mature EDR, SIEM, and identity coverage

• Monitoring firewall traffic logs to detect C2 communications, unusual outbound connections, and unauthorized port activity

• Measuring SOC effectiveness through analyst-driven threat intelligence and direct customer feedback rather than MTTD and MTTR

• Applying micro-macro thinking from military intelligence to widen investigation scope beyond the immediate alert

• Preventing burnout through workload ownership, career-goal conversations, and avoiding process improvements that mask headcount gaps

• Building executive trust through data-driven options frameworks that give leadership decision authority on security investments

• Hiring for non-traditional backgrounds to build analyst teams with diverse problem-solving approaches and thought processes

Listen to more episodes: 

Apple 

Spotify 

YouTube

Sneha Regmi, Director of Security Operations & Resilience Engineering at a major Financial Services organization,  has an incident command framework that prioritizes scope and impact determination over immediate containment, even when executives are panicking. Her teams assign ownership in the first 60 seconds, then the lead verbalizes every decision and next three actions aloud, continuous narration that keeps stakeholders aligned and prevents chaos. She pulls subject matter experts into preliminary investigations early, building credibility to make time-sensitive calls later without second-guessing. 

On insider threat, Sneha flips the standard monitoring-first approach. Her framework starts with prevention controls around business-critical systems, then layers detection only where prevention blocks legitimate work. Prevention without detection leaves blind spots; detection without prevention means everything looks normal until it's not. Her teams renamed the program from "insider threat" to "insider risk" after realizing the original framing damaged organizational trust.

Topics Discussed:

• Assigning incident ownership within the first 60 seconds and verbalizing every decision to prevent stakeholder panic

• Eliminating traditional tiered SOC structures in favor of engineering-enabled responders who write detections and handle incident response

• Prioritizing scope and impact determination over immediate containment to avoid rushing decisions during high-pressure incidents

• Building blameless retrospective practices that enable teams to make split-second decisions without fear during future critical situations

• Implementing prevention-first insider threat frameworks around business-critical systems before layering detection controls

• Pulling subject matter experts into preliminary investigations early to build credibility for time-sensitive containment decisions later

• Managing security operations burnout by setting clear escalation criteria for weekend pages versus business-hours workflows

• Leveraging AI and automation for alert backlog triage while reserving human decision-making for high-impact critical investigations

Allen Carter, former Director of IT Security Operations, ran security operations at Gilead Sciences for a decade, building three teams including a global SOC across India, the UK, and multiple US locations. He developed an approach to burnout prevention where managers function as coaches who spot which team members are "struggling with a twisted ankle" before exhaustion hits,. When onboarding SOAR to automate repetitive alerts, his team saw the technology as a relief, but the implementation work to eliminate false positives nearly burned them out. He learned to celebrate the completion milestone explicitly to maintain morale through the grind.

Allen's incident reporting framework separates security from IT incidents. Whereas IT outages demand "all hands on deck," security incidents require controlled information flow; the wrong person panicking early can trigger a cascade worse than the breach itself. His dual template system keeps operational details within security while board-level reports stay sanitized. He also touches on how, for technology evaluation in pharma R&D environments, less than half of out-of-the-box vendor alerts proved useful. His OT/IoT deployment went operational with massive data volumes that weren't actionable, forcing reactive tuning. Vendor relationships that prioritize understanding your non-commodity threats outweigh feature matrices.

Topics Discussed:

• Building institutional training programs that create visible advancement pathways for SOC analysts beyond graveyard shift roles

• Implementing manager-as-coach models to identify team member burnout signals before exhaustion impacts performance and retention

• Distinguishing security vs IT incident response through controlled information flow versus all-hands-on-deck escalation approaches

• Creating dual incident reporting templates that maintain operational details internally while providing board-optimized communication 

• Evaluating security tech vendors based on relationship quality and non-commodity threat understanding, not feature matrix comparisons

• Managing post-deployment tuning for OT/IoT monitoring to filter unusable industrial control data in operational environments

• Addressing clinical trial security risks where third-party hospital breaches can invalidate months of patient treatment data

• Hiring SOC analysts with deep technical networking knowledge over candidates with security certifications but shallow IT foundations

Listen to more episodes: 

Apple 

Spotify 

YouTube

Robert Maxwell, Security Operations Leader, has a peacetime versus wartime operating model that gives analysts flexibility during normal operations to balance out the 16-hour days that often happen during incidents. He also automated Google Drive "did you share this publicly on purpose?" alerts into Slack bot interactions, eliminating repetitive analyst work.

Robert also amplifies team successes upward and absorbs criticism downward, but scope creep kills incident response teams when executives reassign them to vulnerability management because "the IR team is good at fixing things." He touches on how eliminating entry-level roles destroys the talent pipeline for Tier 2 and Tier 3 and that alert prioritization judgment comes from processing thousands under time pressure. 

Topics Discussed:

• Using "explain how the internet works" interview questions to identify candidates who demonstrate intellectual honesty and research skills 
• Peacetime vs wartime operating models that balance analyst flexibility during normal operations with intensive incident response expectations
• Automating repetitive Google Drive security alerts through Slack bot interactions to free analysts from time-consuming workflow tasks
• Maintaining 8-12 direct report spans of control to enable meaningful people development rather than administrative timecard management
• Preventing scope creep that transforms effective incident response teams into catch-all security functions
• Preserving Tier 1 analyst roles as essential talent pipelines for developing Tier 2 and Tier 3 expertise through alert triage experience
• Building alert prioritization judgment through thousands of real-world investigations rather than skipping directly to complex security work
• Addressing staffing redundancy failures that ignore team vacation patterns and create unsustainable SOC coverage gaps

Listen to more episodes: 

Apple 

Spotify 

YouTube

Joe Albers, Director of Information Security Operations at Element Solutions, Inc., manages a six-person follow-the-sun security team with a counterintuitive framework: accept reduced alert coverage for 6 months while building strategic automation, then gain exponentially more capacity for threat hunting. His approach to AI rejects black box solutions in favor of transparent contextual enrichment that surfaces device background, related tickets, IP ownership, and cross-tool correlations directly to analysts. 

The OT security challenge exposes assumptions that break outside corporate IT. Joe manages environments where patching happens quarterly instead of instantly, and upgrading security tools can void warranties on multimillion-dollar control systems. His vulnerability management shifts from immediate remediation to detailed risk registers documenting why specific vulnerabilities cannot be mitigated and what compensating controls exist instead. 

Topics Discussed:

• Why deliberately sacrificing short-term alert triage for strategic automation buildout creates exponentially more SOC capacity
• How transparent AI augmentation through contextual enrichment accelerates junior analyst development without replacing human judgment
• The leadership framework that treats analyst mistakes as team failures rather than individual performance issues, creating psychological safety
• Why IT security's instant patching model breaks in OT environments where quarterly cycles, warranty constraints, and production disruption risks require risk registers over rapid remediation
• How threat actors consistently exploit holiday periods when SOC staffing drops and response capacity diminishes
• The vulnerability management shift from immediate IT patching to OT risk documentation
• Why hiring for curiosity and basic networking knowledge produces better security analysts than extensive credentials
• How to present security ROI by translating annual tool costs into prevented business losses
• Why succession planning focuses on identifying analysts who actively ask why and how rather than those with the most certifications

Listen to more episodes: 

Apple 

Spotify 

YouTube

Robert Keefer, Associate Director of Security Operations at The Pew Charitable Trusts, has reversed the traditional security equation by building defense in depth that forces attackers to succeed multiple times rather than once. Unlike opportunistic criminal attacks that move on when initial methods fail, nation-state actors now specifically target nonprofits to destroy their ability to gather and disseminate truth, requiring continuous defense rather than single-point protection. His framework combines outsourced SOC providers, automated response systems, and zero trust principles, creating multiple layers where each bypass triggers immediate team response.

Robert positions security as a mission enabler rather than cataloging potential disasters, showing executives how to navigate regulatory requirements like GDPR without disrupting operations. He builds partnerships by being prescriptive about security goals while leaving implementation entirely to subject matter experts, treating each team member as a force multiplier rather than someone to micromanage. The philosophy extends to talent retention through genuine work-life balance where vacation means complete disconnection, mission-driven hiring that attracts people who prioritize purpose over maximum compensation, and vulnerability as a leadership strength. 

Topics Discussed:

• Why nation-state actors now specifically target nonprofits, requiring different defense models than opportunistic criminal attacks
• Building defense in depth that forces attackers to succeed multiple times before reaching valuable assets rather than defending perfectly
• The prescriptive rather than proscriptive security approach that defines goals while leaving implementation to subject matter experts
• How outsourced SOC providers enable continuous level-one triage through hundreds of rotating analysts who stay alert
• Getting executive buy-in by positioning security as a mission enabler that streamlines operations
• Attracting and retaining security talent through mission alignment, genuine work-life balance, and vulnerability as leadership strength
• The shift from passwords to passphrases with MFA that eliminated help desk bottlenecks and half-day downtimes for remote workers.
• Why security leadership has become a people role, with effective leaders spending time away from computers to build partnerships
• The democratization of cybersecurity decision-making as organizations split CISO responsibilities by function and push security decisions down to teams doing day-to-day work

Andrew “AJ” Jarrett, Director of Cyber Monitoring & Incident Response at DTCC, applies emergency response frameworks from his firefighting career to build SOC teams that execute under pressure rather than panic. His approach centers on the Incident Command System, where establishing clear roles, management by objectives, and documentation unit leaders replaces ad-hoc crisis response. Even junior analysts cycle through incident commander roles, building muscle memory for when real incidents strike at 2AM. 

The apprenticeship pipeline in cybersecurity faces an existential threat as organizations rush to replace tier one analysts with AI. AJ identifies this as the critical challenge for the next 5 years, not because automation is wrong but because eliminating entry-level roles breaks the path to developing tier-three analysts and team leads. His interview process prioritizes soft skills over technical certifications, asking candidates about their stress management systems, ethical decision-making frameworks, and whether they have hobbies beyond studying more cybersecurity. 

Topics Discussed:

• How the Incident Command System from emergency response creates SOC teams that execute rather than panic when incidents strike.
• Why reducing signal-to-noise ratios through obsessive tuning matters more than adding new security tools for managing alert volume.
• Interview questions that reveal whether candidates can handle SOC pressure, such as about their support systems and personal ethics.
• Why promoting technical experts into people management roles without leadership development creates more bad managers.
• How PTO as "prepare the others" ensures analysts can disconnect completely rather than remaining tethered to unfinished work.
• The apprenticeship crisis emerging as organizations eliminate tier-one roles without preserving the pipeline for developing senior analysts.
• Why AI analysts need extensive training from senior team members before junior analysts can learn from them without oversight.
• Moving security budget conversations from fear, uncertainty, and doubt toward quantified risk management that executives can evaluate.
• The shift from location-based security models to zero trust architectures accelerated by pandemic-driven remote work adoption.

Listen to more episodes: 

Apple 

Spotify 

YouTube

When seven different responders and law firms were thrown into the same ransomware negotiation chat by a threat actor, Eder Ribeiro, Director of Global Incident Response at TransUnion, it became his framework for managing global incident response: map the data, map the people, and look as holistically as possible before acting. To do this, executive trust must be built long before the 3AM phone call requiring million-dollar decisions. Eder breaks down complex security issues "Barney style" and with empathy, remembering that instruction works best when adapted to how the audience receives it, not how the teacher wants to give it.

For emerging AI risks, he's tracking prompt injection as the attack vector that creates a more linear path to data, particularly in enterprise bundle add-ons that sit in the gray zone between public tools and properly isolated solutions. When investigations spiral, he returns to "control the controllables," reset without finger-pointing, and compress what should take weeks into days. His military-informed leadership philosophy centers on generating agency and freedom for his team, accepting that incident response inherently lacks balance and compensating through daily autonomy.

Topics Discussed:

Listen to more episodes: 

Apple 

Spotify 

YouTube