Securing agents in production: lessons from the frontline
Hosted by Realm Labs + Shivathmika Ramanathan
Register with organizer
What to look forward to
Security leaders from Vercel, Notion, Sierra, and Realm Labs will examine real-world AI agent incidents, how teams assess security tools, and what they build versus buy, ending with practical advice for CISOs. The panel brings expertise spanning product and infrastructure security, fraud and abuse defense, threat research, and a decade of AI security experience, moderated by FastForward publisher Ron Miller.
Inside the event
Enterprises are shipping agents that hold credentials, call tools, and work through complex tasks with minimal supervision. Join us as we delve deep into the problems keeping security leaders up at night, and how they have changed in just the last six months.
Panel
What we'll get into
- Agents in production: where are the real incidents and near-misses happening?
- How are you evaluating AI security tools today, and what makes a vendor credible versus noise?
- Build vs. buy: what are teams solving in-house, and where are they stuck?
- Lightning round: one thing every CISO should do before the end of the year.
Panelists
Mukund Sarma, VP of Security, Vercel - Previously Deputy CISO and Head of Product Security at Chime. Mukund's experience spans the full security lifecycle, from application and cloud security to detection and response, offensive security, and GRC. He also advises and invests in early-stage security startups, particularly in data security and identity.
Jacob Brackett, Head of Application & Infrastructure Security, Notion - Previously Head of Security at Merge, and led enterprise and application security at One Medical. At Notion, Jacob's team is working on agent security as AI becomes core to how millions of people work.
Aditi Gupta, Security Engineering, Sierra - Aditi spent six years at Netflix building large-scale fraud and abuse defenses and four years at Uber developing bot defense and access control systems. At Sierra, she focuses on agent safety, bringing deep expertise in security architecture and distributed systems.
Saurabh Shintre, Founder & CEO, Realm Labs - Previously Sr. Principal Researcher at Symantec and Principal Threat Scientist at Splunk. Saurabh has worked on AI security for a decade. Realm Labs builds runtime control for AI agents and was a Top 10 Finalist at the 2026 RSAC Innovation Sandbox.
Moderator
Ron Miller, FastForward. Former TechCrunch enterprise reporter, now publisher of FastForward, a newsletter on how AI is being put to work in enterprises.
Agenda
- 5:30 PM: Doors and drinks
- 6:00 PM: Welcome from Realm Labs
- 6:05 PM: Panel
- 6:30 PM: Audience Q&A
- 6:40 - 7:30 PM: Happy hour and light bites
Swig
Swig, 561 Geary St, San Francisco, CA 94102, USA. Confirm the venue with the organizer before traveling.
Open in Google Maps ↗Discovered through Cerebral Valley. Details come from the event listing and the organizer page; registration is managed by the organizer.
Last checked: Oct 27, 2026. Confirm current availability before making plans.
Organizer page checked: Oct 27, 2026. Check registration for the latest details.