Foredragsholder
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.
Om sesjonen
Your agent has 100k tokens of context. It still forgets what you told it two messages ago. Prompt "engineering" taught us to craft the perfect instruction. Context engineering treats everything your agent knows as an engineering problem: what it sees, how it retrieves it, what it remembers, and how you prove any of it works.
This talk dissects four antipatterns killing your AI agents and the architectural fixes that actually work:
* The Stuffed Prompt: You crammed everything upfront and hoped for the best. Static context doesn't scale. Dynamic loading and context refinement, fetching what's needed when it's needed, keeps you within your context window without losing signal. And yes, position matters: models do lose track of what's buried in the middle.
* The Wrong Tool for the Job: You picked one retrieval method and used it everywhere. RAG isn't always the answer. Neither are tools. Neither is an exact match. When do embeddings help, when does MCP give you precision, and when does a simple lookup beat both?
* The Goldfish Agent: Your AI agent forgets everything between sessions. Or worse, remembers everything forever. Short-term and long-term memory, pruning and compaction strategies: what to persist, what to summarize, where to store it, and when to let go.
* The Vibes Eval: You shipped because it "felt right." You can't improve what you don't measure. Eval strategies that prove your context choices work — or expose the tokens you're wasting.
Your context window called. It wants its tokens back.
Bonus: Baruch uses a coding agent to demonstrate these patterns live, so you'll see how they work under the hood — but everything applies to AI agents in general.
