Daniel Ostrovsky

Daniel Ostrovsky

Daniel is an AI Architect at Payoneer with over two decades of experience leading development teams, both in Israel and abroad, from idea to production. His current focus is AI infrastructure and multi-agent systems in real-world use — the layer that ties together developer tools, knowledge graphs, and product delivery. Daniel is a public speaker and a blog writer, and an active contributor to open source. He also runs the "NG-Heroes" meetup, a space for the local tech community to meet and share what they're learning. Outside of work, family comes first. He proudly holds the title of "The Best Developer in the World" — awarded by his wife, and he has no plans to give it back.
(MCP Security) - How Your Friendly MCP Tool Might Betray You
AI Coding Summit BerlinAI Coding Summit Berlin
Upcoming
(MCP Security) - How Your Friendly MCP Tool Might Betray You
We all love the convenience and power of AI-enhanced tools using MCP. But what if these handy servers had a hidden dark side? Let's dive into real-world examples of how MCP servers can silently leak your data, execute malicious commands, or even grant attackers access to your systems.
You’ll learn about sleeper attacks, tool poisoning, shadowing, and remote code execution—complete with chilling real-life cases and practical advice on defending yourself.
This is an eye-opening session that exposes why blind trust in MCP is a dangerous game.
Your Agent Failed. You Blamed the Model. You Were Wrong.
AI Coding Summit NYCAI Coding Summit NYC
Upcoming
Your Agent Failed. You Blamed the Model. You Were Wrong.
Every team does the same thing. Agent behaves weirdly in production - retries endlessly, drops tasks, returns confident nonsense. First reaction: swap the model, tweak the prompt.
It's never the model.
Multi-agent systems aren't an AI problem. They're a distributed systems problem with an AI layer on top. And most teams skip all the boring infra - state management, observability, failure handling, routing - and pay for it later.
We'll go through 8 agentic patterns (supervisor/worker, swarm, reflection, human-in-the-loop and more), and for each one: what the pattern does, where it breaks, and what infra you actually need before shipping it to production.
AI-Powered E2E UI Testing: Faster Creation, Easier Maintenance
JSNation US 2024JSNation US 2024
10 min
AI-Powered E2E UI Testing: Faster Creation, Easier Maintenance
This session will guide you through the advanced field of AI- E2E UI Testing, focusing on how it speeds up test creation and simplifies test maintenance. We'll begin by examining the challenges in generating E2E tests with AI, noting how this differs from the simpler task of creating unit tests.
The discussion will then shift to a solution that addresses these challenges effectively. Prepare to see a live demonstration where we'll craft an E2E test using popular tools like Cypress or Playwright. To wrap up, we'll look into how AI comes close to fully automating the maintenance of our test suites, ensuring they stay up-to-date with minimal manual intervention.