Notable verbatim quotes

Topic tags in brackets. Line ranges refer to transcript.md.

1. [problem framing] "It's 2026. Why are you overviews and code reviews are still on like an engineering team?" — §1, opening.
2. [trust deficit] "People exclaim that they don't really rely on the agent to close the feedback control." — §2.
3. [agents ignore specs] "they have specs. And they see that the coding agent multiple times just ignore most of the specs part of it." — §2.
4. [motivating example] "the continue button is exactly where I asked him not to put it." — §3.
5. [diagnosis] "coding agents that we are using are not focused on verifying features. They are amazing in generating code" — §4.
6. [code-vs-code isn't enough] "comparing code to code isn't the answer. You need to deploy the feature in your staging environment and really see how it looks like." — §4.
7. [single-agent fails] "One agent can take all the requirements that are in the ticket, all the requirements that are in the specs, all the design aspects… and also go and verify them. It just explodes the agent." — §5.
8. [planner/verifier split] "Planner's role is to extract the requirements from the spec and understand what are going to be the failure cases… Only one task to extract requirements." — §6.
9. [sub-agent delegation] "instead of one agent that is checking 10 or 12 or 15 requirements sequentially, I'm going to have 15 or 12 agents that are running in parallel. Each of them is reaching to a specific verdict and in the end there's an orchestrator that collects all the verdicts." — §7.
10. [grounding] "The specs are a snapshot and what we're trying to achieve. In this coding task. But the code is how we ground that agent to reality." — §8.
11. [base branch over diff] "if we're giving it the diff. It's biased to the specific solution that the engineer choose to implement. But if we give it the base branch before the change, the agent is open-minded to different kinds of approach and is more critical about the solution that was chosen." — §8.
12. [scoping] "if I'm reviewing a front end feature, there's no reason to be concerned about backend issues because I'm just going to create noise that are irrelevant for this specific feature." — §8.
13. [sandbox rationale] "clicking on unknown URLs, it's just like clicking on phishing link that I'm getting on an SMS on an email. And I'm going to do it a hundred times. A month." — §9.
14. [ephemeral sandbox] "we used together with AWS agent core… the ability to run an ephemeral sandbox for every requirement that is being validated." — §9.
15. [key takeaway 1] "It's 2026. And a context engineering is still a hard problem." — §11.
16. [key takeaway 2] "you have your specs and you have your code… if you combine them together, it's really a gold mine." — §11.
17. [key takeaway 3] "if you're entering an area that you think is a dangerous area related to cybersecurity, vulnerabilities prefer using a third party tool instead of developing yourself." — §9 / §11.
18. [strategic advice] "Look for the gaps that the big coding agents are not able to fill and build the product there." — §11.
19. [tests vs spec verification] "the problem that I have with tests is that they test things that nobody really… they're not testing real life. Like QA teams are sitting and thinking about these imaginary scenarios that are not going to be anytime in your product." — §13.
20. [renewable energy byproduct] "when they knew that the specs are used to be for the agents to verify the feature, it encouraged them to write better specs. So it's like renewable energy" — §13.
21. [model selection] "For extracting verification, the requirements we're using a heavy model… While the verification ones, we used like small agents" — §14.
22. [focus where it matters] "we'll be focusing on the one or 5% in the product that matters most. And not going into every part in the product" — §14.