Baruch Sadogursky (@jbaruch) did Java before it had generics, DevOps before there was Docker, and DevRel before it had a name. He built DevRel at JFrog from a ten-person company through IPO, co-authored "Liquid Software" and "DevOps Tools for Java Developers," and is a Java Champion, Microsoft MVP, and CNCF Ambassador alumni.
Today, he's obsessed with how AI agents actually write code. At Tessl, an AI agent enablement platform, Baruch focuses on context engineering, management, and sharing. On top of sharing context with AI agents, Baruch also shares knowledge with developers through blog posts, meetups, and conferences like DevNexus, QCon, Kubecon, and Devoxx, mostly about why vibecoding doesn't scale.
Software is a mess. Everything feels bloated and sluggish. Bugs pile up. Updates break more than they fix. Even basic apps demand absurd amounts of computing power. And innovation? If you count shuffling UI elements or slapping a ChatGPT button onto everything, sure.
The problem isn’t bad engineers. It’s that building software has turned into an obstacle course. Every step—writing, reviewing, deploying—is buried under layers of tools, approvals, and abstractions. A simple change takes weeks. We spend more time wrangling complexity than shipping code. No wonder quality is tanking, and real progress feels nonexistent.
So how did we get here? More importantly, how do we stop making it worse? Fixing the Googles and Metas of the world might be out of reach, but your projects don’t have to follow the same path. This talk is about cutting through the mess—how AI, smart tooling, and better developer workflows can actually help instead of adding more noise. Let’s figure out how to build good software again.
We are witnessing a fundamental shift in software engineering. As AI
agents become capable of generating entire features, the bottleneck is
no longer writing syntax—it is defining intent. This evolution brings
Spec-Driven Development (SDD) back into the spotlight, not as a
bureaucratic documentation step, but as the primary interface for
software creation.
In this panel, we explore the reality of building software when you're
pair programming with an LLM. We will discuss the architectural
implications of moving from imperative coding to declarative
specifications. How do we define unambiguous requirements for
non-deterministic agents? Does "prompt engineering" evolve into rigorous
system design? Join us to discover if the Spec is truly the new Source
of Truth, or just another layer of abstraction to maintain in a world
where code is becoming a commodity.
Searching for speaker images...