The Speed of Context: Why AI Changed What Engineers Actually Do

Most engineering teams are still optimizing for the wrong thing. They chase the speed of code when the real bottleneck is the speed of context. Matt Watson and Eban Bisong, founder and CEO of Senvi, get into what actually changes when AI moves from a coding tool to a teammate.

Eban has spent his career as a founding engineer, and his approach is hands-on: don't tell skeptical engineers AI works, show them, every standup, until the pushback turns into excitement. At Park DNA he built "RTD2," an OpenClaw-powered droid wired read-only into their data sources, Slack, and Jira. It answered support questions before an engineer could, created its own bug tickets, and joined meetings through Fireflies so nothing got lost. The lesson underneath all of it: record everything, because the team that captures the most context ships the right thing fastest.

Matt also shares his own three-week rabbit hole with Claude Cowork, $8K in tokens, a fully rebuilt Full Scale website, a thousand dead blog posts deleted, and 200 more rewritten. They go a few rounds on why it's a bad time to be a coder but a great time to be a builder, why "good enough" is a real standard and not a cop-out, and why ownership beats asking permission every time.

If you build software or lead an engineering team, listen now. And if you want to try Eban's voice-first AI journal, visit senvi.ai.

⏱️ Episode Breakdown

00:42 From Founding Engineer to Solo Founder

01:52 Using AI as an Engineering Leader

03:24 Building RTD2: An AI Teammate for Support

05:27 The Speed of Context, Not Code

06:11 Why You Should Record Everything

08:59 Winning Over AI-Skeptical Engineers

11:50 The AI Spectrum Across 80 Clients