High Signal Podcasts Evidence ledger
Method
Browse
← Back to evidence

Evidence receipt / belief

Published · transcript-backed

Alexander Embiricos: belief

14 Dec 2025 Lenny's Podcast Why humans are AI’s biggest bottleneck (and what’s coming in 2026) | Alexander Embiricos (OpenAI Codex Product Lead)

“" You would start by first making sure they understand the code base and then you would maybe align on an approach and then you would have them go off and do bit by bit. And I think if you use Codex in that way, you'll just naturally start to understand the different ways of prompting it because it's a super powerful agent and model, but it is a little bit different to prompt Codex than other models.”

— Alexander Embiricos

Source trail

Everything needed to verify it.

Speaker
Alexander Embiricos
Attribution
Verified speaker
Claim type
belief
Recorded
14 Dec 2025
Publisher
Lenny's Podcast

Transcript context

…If someone was just getting started, is there a tip you could share to help them be successful? If you could just whisper a little tip into someone just setting up Codex for the first time to help them have a really good time, what's something you would whisper? I might say try a few things in parallel. So you could try giving it a hard task, maybe ask it to understand the code base, formulate a plan with it around an idea that you have and kind of build your way up from there. And the meta idea here is, again, it's like you're building trust with a new teammate. And so you wouldn't go to a new teammate and just give them like, "Hey, do this thing. Here's zero context. " You would start by first making sure they understand the code base and then you would maybe align on an approach and then you would have them go off and do bit by bit. And I think if you use Codex in that way, you'll just naturally start to understand the different ways of prompting it because it's a super powerful agent and model, but it is a little bit different to prompt Codex than other models. Just a couple more questions. One, we touched on this a little bit, as AI does more and more coding, there's always this question of, "Should I learn to code and why should I spend time doing this sort of thing?" For people that are trying to figure out what to do with their career, especially if they're into software engineering computer science, do you think there's specific elements of computer science that are more and more important to lean into, maybe things they don't need to worry about? What do you think people should be leaning into skill-wise as this becomes more and more of a thing in our workplace?…

Stored transcript either side of the excerpt. The highlighted words are the published quote; the surrounding text is unedited source, never generated.

Search evidence