Connect your Claude, Codex or OpenClaw for full transcripts on millions of podcasts. Two week free trial, cancel anytime.
The Analytics Engineering Podcast · Hands-On Engineering Podcasts · October 2026
Shankar and Tristan Handy discuss the flood of highly personalized, AI-written cold email, which spam filters struggle to tell apart from a real person writing.
…like making this instruction a good prompt injection. Like I think that the AI agents ignore it. But my guess is that because I have this instruction, the AI agents that are just trying to find professors and mass are probably not crafting an email for me. And I'm very, very happy with that. So
…is like highly personalized, clearly the result of like, you know, agentic context development on you personally before sending. Like, it's, I don't know how kind of traditional anti-spam techniques would combat that. It's not that different from like an actual human sending you an actual email.
Right. I think it's super fascinating. I actually tell people now, like, just don't use AI to write emails and you will get noticed. Like, it's, it's never been easier to get a job or to get noticed now because all you have to do is not use AI. I think it's kind of ironic.
So you mentioned your brand new CMU job. Maybe you could just spend a few minutes talking about your history and how you got to where you're at.
Yeah. So I went to college. I majored in computer science. I did my undergrad at Stanford. I did this during the time of the deep learning boom. So 2016 era, like it was right before Transformers, but everyone was like, you know, Gantt, adversarial image classification. I don't know, like this was…
Did you go to Neurops?