Beyond the Black Box: Building AI Agents That Scale

Join us for a meetup exploring both the technical foundations of AI coding agents and the practical lessons from rolling them out across a large engineering organization. The event will be held in Hebrew Agenda: 18:00 - Gathering, mingling, food + drinks 18:30 - Not a Tooling Problem: Lessons Learned from Org-Wide Agents Rollout - Adam Yahid - Software Engineer Team Lead, monday.com 19:00 - Go Deep Inside the Harness → Make Better Decisions - Yoni Goldberg, Independent Consultant: SDLC · Testing · Architecture 19:30 - Break 19:45 - Company Brain with Microsoft Fabric - Roey Zalta\, Data & AI Solution Engineer\, Microsoft Abstracts: Talk 1: Not a Tooling Problem: Lessons Learned from Org-Wide Agents Rollout - Adam Yahid- Software Engineer Team Lead, monday.com Rolling out AI agents across an organization sounds like a technology problem, but it isn't. It's a change management problem with a technology component, and the difference shapes every decision you make. In this talk, I'll walk through how we built and deployed a developer on-call agent across monday.com's engineering org, what actually worked, and the five lessons we learned about what we'd do differently, or the same if we started over today. Talk 2: Go Deep Inside the Harness → Make Better Decisions - Yoni Goldberg, Independent Consultant: SDLC · Testing · Architecture Your team is in a meeting room, AI coding decisions on the table. Someone says: "A reviewer sub-agent can chat back and forth with the main agent." "The deny list blocks bash scripts deterministically." "Few hierarchical skills beat lots of short ones." True? False? Most people aren't sure. The real answer is: it's complicated - and that complexity shapes the decisions you make. These questions have real answers, and the answers live inside the harness. This talk is a seriously deep journey into how your harness works - Claude Code, Cursor, whatever you run. Once you see it, the black box turns to glass - and every decision you make gets sharper. Love code-level talks? This one is for you Talk 3: Company Brain with Microsoft Fabric - Roey Zalta - Data & AI Solution Engineer, Microsoft AI automation is no longer blocked mainly by model quality. The bigger challenge is that domain knowledge is trapped inside the organization: in documents, tickets, meetings, chats, tables, decisions, and business processes. In this 30-minute session, we will explore how to build a “Company Brain” with Microsoft Fabric: a living semantic graph that turns organizational knowledge into memory agents can use. We will cover the core ideas behind a Company Brain, including ontology, knowledge graphs, persistent factual and decision memory, role-based lenses, governance, and integrations with tools such as Jira, GitHub, Teams, Slack, Confluence, and Gong. The session will focus on one practical question: how do you start small? We will look at how to choose one high-value use case, define a first MVP architecture, and connect it to OneLake, Copilot, governance, and the Fabric platform. By the end, participants will have a clear mental model for moving from scattered enterprise knowledge to an agent-ready Company Brain.