preston-kwei

Having attended startup school this weekend, we were told how, in the past, companies were reluctant to try new technology from startups, AI products, etc., because they were unvalidated and unsure of how they would work. Now, companies are more worried that they're going to be left behind and that their competitors will beat them if they do not adopt these new technologies.

I'm curious about how the LLM-based detector worked. How often did you run into a false positive? (I'm assuming you leaned on the more sensitive side due to the importance of the data)

Do you think customers trust it either? I'm trying to differentiate between "good enough" and when enterprise customers actually care. I mean, it doesn't matter until a customer asks for their data.

I agree. I'm trying to sell to enterprise right now and this is a blocker. Basically, my product stores emails in S3, but metadata in Supabase, and caches in Redis. It's all fragmented and super difficult to keep track of. Plus Supabase isn't WORM compliant so it is hard to ensure data doesn't get deleted.

I think UI quality is going to stop being as important when anybody can just generate an "average" UI that is good enough in minutes.

Ultimately, this really just shifts the focus towards product design and ideation rather than UI design.

How many emails does it usually take for you to lock something in? And do you end up following up manually if they don’t respond?

How much of this issue is availability vs just getting them to respond and commit to the meeting?

I think the back and forth is one of the most time-consuming parts. How do you manage it?

Interesting. Do you end up just accepting the meetings or do you try to push back?

I’m more curious how this impacts trust than anything else.

In the span of basically a week, they accidentally leaked Mythos, and then now the entire codebase of CC. All while many people are complaining about their usage limits being consumed quickly.

Individually, each issue is manageable (Because its exciting looking through leaked code). But together, it starts to feel like a pattern.

At some point, I think the question becomes whether people are still comfortable trusting tools like this with their codebases, not just whether any single incident was a mistake.

Thank you for your response. I think the ICP needs some refinement. I've found it pretty hard to prospect because it is so broad.

> People generally don't have problem with email scheduling I think this stems from the fact that there is no existing tool that solves this problem. Clearly, tools like Calendly haven't solved it for execs or people with complex scheduling flows. A lot of it is caused by the fact that people tend to just tolerate scheduling because they don't know a better way, leading them to waste more time than they expect.

> The moment Persona is compared to a human EA, you lose on judgment, context, and trust. I agree with this. I'm trying to steer away from this actually, because a lot of companies sell an AI EA, even though their product doesn't even do anything related to EA tasks.

I think the landing page needs more work, copy wise. I'm certainly moving towards the control thing. The features are in place, so I need to tweak our marketing.

> Accept that this may not be a standalone company Totally. Scheduling alone cannot be a standalone company. That's just the first pain point I'm attacking. I have many more features that are in progress that move it more into it's own category. Think of it more as a email delegation layer for AI agents.

Thanks again for your thoughtful response.

I'm not sure how AWS logs work since I'm just used to looking at Cost Explorer... But I would assume they actually store the history themselves, and you can just pull it.

Fair, I think I'm certainly in the minority. Especially now more then ever with an increasing amount of non-technical people exploring vibe coding, 'good enough' really is good enough for most users

Yes on the satire. This post made me laugh :).

I agree with this framing a lot, especially the idea that judgment is the bottleneck

In my experience building Persona, an AI scheduling assistant, the most useful role for humans isn't to be always in the loop. LLM's are terrible at making judgement calls, especially when the right choice depends on a specific user's priorities and the confidence is low. However, even with low confidence, the llm still needs to make a guess.

I think an interesting use case for this would be to have llm's be able to ask questions to users when they hit a specific level of uncertainty. These could be directly answered by a human, or inferred as the user uses the product more.

That feels more scalable than completely blocking human-in-the-loop queues and more honest than pretending the model already knows the user’s preferences.

Well unfortunately, that’s just how I write. None of my posts are LLM-generated, so I'm sorry they come across that way.

Historical tracking is useful, but only if it’s dead simple. If it requires standing up a DB or service, I probably won’t use it.

I've been trying to track down cloud waste recently after realizing EC2 was storing 10GB snapshots every 12 hours for the past 8 months and I didn't realize. Obviously not a crazy 20k bill, but still annoying -- especially at small scale.

While this may be true for casual users, for dev native products like Codex, the desktop experience actually matters a lot. When you are living in the tool for hours, latency, keyboard handling, file system access, and OS-level integration stop being “nice to have” and start affecting real productivity. web or Electron apps are fine for experimentation, but they hit a ceiling fast for serious workflows -- especially if the icp is mostly technical users

Thanks for your thoughtful reply. I really like that quote -- hopefully I start another company and think about distribution straight off the bat. Can I ask what the product was? Even just a general idea. And, how did you find the emails.

I've found myself procrastinating a lot when it comes to trying to acquire users. It's a lot more fun to build features than it is to do cold outreach. Yes, you found the site :).

1) Understandable, let me reply to the other 3 points to see if pricing makes sense after that.

2) We’re not trying to replace Calendly. Calendly is great for link-based 1:1 scheduling, but many execs still operate primarily over email and actively avoid sending links. Persona is an email delegation layer that handles inbound scheduling end-to-end: proactive follow-ups, coordinating across long threads, and dynamically rescheduling when higher-priority contacts come in. That distinction should definitely be clearer on the landing page.

3) Fair point that you’re not the audience. We’re focused on founders and execs who spend a meaningful amount of time coordinating meetings over email. For them, the pain isn’t finding a time once, it’s the constant back-and-forth, follow-ups, and reshuffling priorities. Not everyone can afford a human EA, and link-based scheduling tools can’t handle that level of personalization (talking to an investor is very different than a friend). Research we’ve seen shows execs still spend 6+ hours per week on scheduling.

4) That’s a good point, and we’re thinking along similar lines. Since Persona operates directly over email, there’s a natural distribution loop there, similar to Calendly links, and we plan to lean into that more deliberately. Would you be bothered if you were on our 23/mo plan and still had a banner there? Also, do you think there should be a free tier beyond the 14 day trial?

Thanks again for taking the time to respond.

Interesting! I bet now with AI, you can create an agent to do this. I'm glad that worked, however, we're targeting founders of b2b saas companies who are early stage, and that is quite a broad search.

The “lump of cognition” framing misses something important. it’s not about how much thinking we do, but which thinking we stop doing. A lot of judgment, ownership, and intuition comes from boring or repetitive work, and outsourcing that isn’t free. Lowering the cost of producing words clearly isn’t the same as increasing the amount of actual thought.

Got it. That's fair, I could definitely remove the first sentence. However, I still want the email to seem personal. Do you think it is important to have that personalized part somewhere in the email?

There's no re, I just put that there instead of saying Subject:. Sorry, that's confusing.

I get the concern if the value is framed as “replacing an admin assistant”, but for many founders, the real win isn’t salary savings, it’s eliminating context switching, follow-ups, and mental overhead around coordination. As well as eliminating the mistakes that can be made by humans. Even small logistics can disproportionately drain attention. If done well, this is less about cost reduction and more about reclaiming focus and execution speed.

Very low, like 0.01%. And oftentimes these people just say no. I try following up, but that usually doesn't do much.

Totally. Here is what my emails typically look like:

Re: quick scheduling question

Hi [name],

I’ve followed [company] for a while and the scale you’re operating at is impressive. With customer and partner meetings, plus internal meetings, I imagine your calendar gets pretty full.

I’m working on a way for founders to cut down the email back-and-forth of scheduling without hiring or managing an EA.

Is scheduling something you still handle personally at all?

– Preston