
SPEAKER
Brandon Waselnuk
DevRel - Unblocked
Brandon Waselnuk is Developer Relations at Unblocked, the context layer for agentic software development. He previously led product at Mintlify and co-founded Squire AI, an AI code review platform. A Y Combinator alum (S21) and Venture Partner at Pioneer Fund, Brandon also led product initiatives at IBM, helping create Design Thinking @ IBM and managing portfolio planning for IBM's $1B+ Business Analytics division. He's passionate about building tools that make engineering teams more productive.
Upcoming sessions
See complete program4 November
4:15 PM · 30 min
Emitted, not authored: The context layer under your software factory
It's your first week on the team. You read everything attached to your first ticket: the README, the wiki, the Linear project. You ship the PR. The review comes back from a senior engineer: "we stopped doing it this way in March, the vendor doesn't accept that shape anymore." Nobody wrote that down. It lived in a Slack thread from an incident two months ago and in the heads of the people who were up that night fixing it. You learned it by getting it wrong.
That's how most engineering knowledge gets earned. Not from documentation someone sat down to write, but from the act of the work itself: the pull request comments, the chat thread where a feature got argued out, the incident channel while people are still responding, the support channel where a customer explained what they actually needed.
I call this emitted context. It was just typed, by people, doing their jobs.
Authored context is what the team decides matters: the rules and skills files for agents, the wikis and READMEs we curate. High leverage, worth doing well. Emitted context is what the team told each other while doing the work. Your agents today are you, the new hire, on every single task, and they never get to build tribal knowledge.
Letting them build it is a lot harder than connecting a Slack MCP. Emitted context contradicts itself, goes stale overnight, and includes things not everyone is allowed to see. This talk is about the context layer that handles that: surfacing context nobody authored, reconciling it with citations, keeping it fresh, with no human in the loop. Automate a factory on authored context alone and it can only act on what somebody remembered to write down, and no one person knows everything the team knows.
I'll run the same task with and without emitted context, show why agents don't stop to ask for what they're missing, and cover the three things we got wrong building this. You'll leave with open source tools we extracted from our context layer, and a clear read on whether your agents can unblock themselves or if you're trapped in the factory.