The Engineered-Mind Podcast

The episode contains some useful framing - agentic AI as a 'horizontal tool' akin to statistics in engineering, and the benchmarking point for evaluating AI use cases - but is heavily padded with generic AI discussion, biographical throat-clearing, and truisms that any informed engineer already knows. Novel ideas-per-minute is low.

“generative AI, agentic AI, these are horizontal tools. These are things that I think will have an impact similar to the use of statistics in engineering or the use of math and engineering”

“you need to have a good benchmark or a good way of measuring what success is going to look like before you start an evaluation”

The framing of engineering tools as guardrails for agentic AI is a mildly fresh angle, and the firmware/training-data point is a small but useful nuance; otherwise the episode recycles standard takes - AI augments don't replace, democratization is good, security is a barrier - without genuine contrarian or first-principles arguments.

“those tools, you know, they define the workflows. They also, from a generative AI point of view, they kind of implement guardrails”

“I think in a way that's much different from previous technology changes where it required more of a fundamental shift and a paradigm shift”

Seth DeLand is a legitimate 15-year MathWorks practitioner with hands-on product management responsibility for generative AI and a genuine engineering background in combustion/controls; he is not a hollow thought-leader, but his seniority and breadth of novel practitioner experience is mid-tier rather than exceptional.

“I've been at MathWorks for just over 15 years in a number of different roles, starting from tech support and on the front lines”

“We have a Matlab co-pilot that's been out on the market. We have a Simulink co-pilot that's on the way. Another thing that we released in November last year was a MATLAB MCP server”

There are a handful of concrete product references - MATLAB MCP server released November, AI Chat Playground built in 2023, Simulink Copilot in beta on GitHub - but no customer metrics, performance numbers, dollar figures, or named case studies; the automotive workflow example is explicitly kept vague and qualitative throughout.

“we released in November last year was a MATLAB MCP server. So MCP is Model Context Protocol”

“Back in 2023, we built what we call the AI chat playground”

The host makes some genuine attempts to probe - asking whether democratization has downsides, what engineers get wrong, and what makes a good use case - but rarely follows up on vague answers, allows several generic responses to stand unchallenged, and frequently editorializes rather than pressing the guest for harder specifics.

“From an engineering point of view, because now AI has been commoditized to some extent, right, and democratized. Do you think that's always a good thing or are there also some negative sides to it?”

“What do you think most engineers get wrong still about AI?”