AI Security Night
Hey everyone,
We’re back with a new edition of the AI Security Engineers Meetup in London on June 23rd with 2 great talks!
Talk #1 - Architecting Safe Autonomy: High-Stakes Autonomous Agents Need Deterministic Checkpoints for safety
In the rush to build autonomous agents, we face a fundamental tension: the more freedom we give an LLM to solve complex problems, the more likely it is to drift into “stochastic hallucinations” or policy violations. Most safety efforts focus on restricting the model via prompts, which often stifles the very reasoning capabilities required for sophisticated tasks.
This talk introduces InsideOut, a design pattern that prioritizes agent autonomy by implementing deterministic checkpoints. Rather than micro-managing the agent’s “thoughts,” the InsideOut architecture allows the agent to navigate freely through a task—provided it periodically “grounds” its progress in structured, verifiable artifacts such as JSON or Markdown.
We will walk through a real-world application of this method: an agent designed for setting up and managing cloud infrastructure. The agent is given the autonomy to discuss and decide on Features, Stack Components, Configurations, Cost Estimates, Terraforms, Deployment, and Management. To ensure reliability, each stage requires the generation of a JSON artifact validated against deterministic rules. If a boundary is breached, the agent is triggered to repeat that specific stage until the output is within bounds, preventing error propagation.
By drawing a parallel to the concept of a Brownian Bridge versus Brownian Motion, we illustrate how these checkpoints act as “pins” that anchor a random walk. This approach demonstrates how forced artifact crystallization allows an agent to design and deploy complex stacks with high independence and zero “drift.” This session provides a framework for building agents that are more productive because they are safely unconstrained.
Speaker: Hossein Kakavand. Hossein did his Ph.D. at Stanford University. He has been with several start up in AI, ML and Distributed Systems, with IPOs on NASDAQ and LSE. He is currently a Co-Founder of Luther Systems focused on solving the Enterprise Operations problem at scale.
Talk #2: Agentic Development Security: Securing the AI Workforce
AI agents are rapidly becoming part of the software development lifecycle, writing code, interacting with tools, and making decisions with increasing autonomy. But as organizations embrace agentic development, a new set of security challenges emerges: How do you trust the tools agents use? How do you prevent dangerous actions? And how do you ensure the code they generate is secure?
In this session, we'll unveil and demo Snyk Agentic Development Security, a new approach to securing both AI builders and the AI-powered tools they rely on. You'll see how organizations can secure the agent supply chain by vetting MCP servers, plugins, and external tools before agents interact with them; govern agent behavior through real-time policies that prevent destructive actions with Agent Guard; and ensure trusted output by automatically detecting and fixing vulnerabilities in AI-generated code at the moment of creation.
Join us for a live demonstration of how security can keep pace with the rise of autonomous development, enabling teams to innovate confidently while maintaining control, trust, and resilience across their AI workforce.
Speaker; Denis Kent - AI Security Engineer at Snyk, focusing on emmerging technologies and solutions
----------------
As always, expect great talks, free pizza and drinks, and even better company.
📍 Where: Snyk Offices, 24 Eversholt Street, London
(Euston tube station is right across the street)
📅 When: June 23rd
## Agenda
6:00 PM – Doors open
6:00 PM – 6:30 PM – Food & drinks
6:30 PM – Talks begin
7:30 PM – Talks end & social time
8:30 PM – Doors close
Looking forward to seeing you there!