Marcel Velica

Marcel VelicaOct 20, 2026

Most companies are rolling out AI agents the way they rolled out shadow IT in 2012. 𝐅𝐚𝐬𝐭. 𝐄𝐱𝐜𝐢𝐭𝐞𝐝. 𝐖𝐢𝐭𝐡 𝐚𝐥𝐦𝐨𝐬𝐭 𝐧𝐨 𝐯𝐢𝐬𝐢𝐛𝐢𝐥𝐢𝐭𝐲 𝐢𝐧𝐭𝐨 𝐰𝐡𝐚𝐭 𝐭𝐡𝐞 𝐚𝐠𝐞𝐧𝐭𝐬 𝐚𝐜𝐭𝐮𝐚𝐥𝐥𝐲 𝐝𝐨. For the past few weeks I've been running with Zenzap parts of my own business on agents: inbox triage, CRM updates, payment reconciliation. I set up one agent and connected my Gmail and CRM. It took 90 seconds, with no code.  I told it what I wanted, the way I'd brief a coworker. 𝐖𝐡𝐚𝐭 𝐢𝐭 𝐝𝐨𝐞𝐬 𝐧𝐨𝐰: My setup isn't one person talking to one bot. It's a shared chat where I work alongside three agents: → One runs deals: labels every thread, flags where I owe a reply, drafts follow-ups. 30 so far, none sent without me. → One runs finance: cross-checks payments against the CRM. It found collaborations I had never logged. → One coordinate: checks the other two's work and tells me only when something needs a decision. My inbox has been at zero every day since October. 𝐓𝐡𝐞 𝟓 𝐟𝐞𝐚𝐭𝐮𝐫𝐞𝐬 𝐭𝐡𝐚𝐭 𝐦𝐚𝐝𝐞 𝐢𝐭 𝐰𝐨𝐫𝐤: 1)Agents inside the chat. My agent works in the same conversation as me, not in a separate tool. 2)Approval cards. Sensitive actions wait for my tap before they run. 3)Scheduled workflows. Recurring jobs like the morning follow-up check and inbox triage run on a schedule I set. 4)Memory. It remembers my decisions and rules, so I don't have to re-brief it every day. 5)App connections. Gmail, my CRM and my calendar, with access limited to what I choose. 𝐀𝐬 𝐚 𝐬𝐞𝐜𝐮𝐫𝐢𝐭𝐲 𝐥𝐞𝐚𝐝𝐞𝐫, 𝐈 𝐜𝐚𝐫𝐞𝐝 𝐚𝐛𝐨𝐮𝐭 𝐜𝐨𝐧𝐭𝐫𝐨𝐥 𝐦𝐨𝐫𝐞 𝐭𝐡𝐚𝐧 𝐬𝐩𝐞𝐞𝐝: 1) Visibility: the agent works in a chat where I can see every action 2) Approval: I decide what it can access and what needs my sign-off 3) Accountability: when something's missed, I know which task and when Agents don't just need to be capable. They need to be governable. If you're bringing agents into your team without losing control, start here: https://bit.ly/try-zenzap

Marcel VelicaOct 20, 2026

ZERO TRUST ARCHITECTURE: FROM USER IDENTITY TO PROTECTED RESOURCES Your network can be secure and still be one compromised account away from a breach. That's why traditional perimeter security isn't enough. Zero Trust changes the approach: → Never trust implicitly. → Always verify. → Grant least privilege. Here's how a modern Zero Trust architecture works: 1. Verify Users ↳ Employees, admins, and vendors ↳ Verify every access request 2. Check Devices ↳ Validate patches and encryption ↳ Check device health and compliance 3. Verify Identity ↳ Use Microsoft Entra ID or Okta ↳ Centralize authentication and SSO 4. Strengthen Authentication ↳ Enforce MFA and passkeys ↳ Block unauthorized login attempts 5. Evaluate Access ↳ Check location, identity, and risk ↳ Apply context-based access policies 6. Make Policy Decisions ↳ Policy Engine evaluates each request ↳ Allow, deny, challenge, or revoke 7. Enforce Access ↳ Policy Administrator applies decisions ↳ PEP blocks unauthorized requests 8. Isolate Workloads ↳ Segment apps, databases, and APIs ↳ Limit lateral movement between systems 9. Protect Resources ↳ Secure cloud, SaaS, and internal apps ↳ Restrict access to sensitive data 10. Monitor Continuously ↳ SIEM and XDR detect suspicious activity ↳ Analytics identify unusual behavior 11. Build a Security Foundation ↳ Encrypt data and prevent data loss ↳ Prepare for incidents and recovery Zero Trust isn't a product you install. It's a security strategy that continuously verifies access and limits what an attacker can reach. If one employee's account were compromised today, how much of your network could an attacker access? Share this with your network if you found it useful, and help more security professionals understand Zero Trust. Reshare to help others strengthen their security knowledge. ➕ Follow Marcel Velica for more practical cybersecurity insights, frameworks, and security architecture breakdowns.

Marcel VelicaOct 21, 2026

Most security decks bury the decision. They're packed with charts and acronyms. And somehow, after twenty slides, someone still has to ask: “So what do you actually need us to approve?” I've put enough security updates in front of executives to know the hard part isn't finding another chart or making an ugly slide look expensive. It's deciding what the story is before you start designing. That's the actual job. So I tried something different. I gave Gamma a rough cybersecurity board brief. The idea was simple: explain what risk changed, why it matters, and which decision leadership needs to make. Gamma 𝟓 approaches that as a visual storytelling agent instead of expecting one magical prompt to produce the perfect deck. It can build the outline before designing. Finally, some restraint. I kept the story to five slides. One message per slide, no walls of text. Because real business decisions deserve something better than a dashboard wearing a PowerPoint costume. And if you prefer the old workflow, you can switch back to Gamma Classic without losing anything you've already created. The work stays, even if the way you build it changes. What's the deck you dread making most? #GammaPartner