Ashir Murtaza

How much control do teams have over each step when the agent starts refining outputs on its own?

I really like this update prompt iteration usually feels messy because you end up rewriting things that were already working fine.

I like the idea of being more intentional with caffeine instead of just reactive.

Cool idea, but how do you prevent manipulation or bot activity?

Social proof only matters after I already have some interest. It won’t make me click, but it might reinforce the decision.

The point about personal profiles outperforming brand accounts is so true, especially on LinkedIn. People connect with people, not logos.

This is why having a “launch relay team” makes so much sense. Even 2–3 people rotating can keep energy and response quality high.

Really interesting that runtime monitoring came up so much in feedback. Feels like there’s a big opportunity there long term, even if your current focus on pre-execution security makes sense.

That’s a bold way to test both the product and the audience. I like the honesty here.

Most platforms force you into one setup, so the “your data, your way” approach stands out. Curious to see how smooth the switch between local and cloud actually feels.