Connect with the Product Hunt community

Hi everyone!

Containment rate (the share of conversations the bot handles without a human) is the default AI support metric. We think it rewards the wrong thing. A bot can contain a conversation by never letting the customer reach a person.

With customers, we track the outcome instead: was the payment made, the policy renewed, the meeting booked, the issue resolved? One real estate customer cut outreach costs by about 70% and finished a month of calls in under a tenth of the time. That's a number a CFO cares about. Containment isn't.

Our AI employees have names. Peter is an SDR, Kavya sells insurance, Arjun is a loan officer, Priya handles renewals.

We did it because customers found it easier to talk about "Peter's pipeline" than "the lead qualification workflow." Teams started treating the agents like colleagues with a job, which made it easier to define what each one owns.

A few of you asked at launch how QuickQuill handles several people. It can now tell speakers apart, on your Mac.

Telling speakers apart (experimental)

Turn on "Identify speakers (experimental)" in Settings General. After a new recording ends, lines are split by voice into Speaker 1, Speaker 2... for calls, in-person meetings on one mic, and imported files. Rename a speaker once and all their lines follow.

It can merge or miss people who rarely speak. Feedback welcome.

We teamed up with @GoogleCloud for a Cloud Run Hackathon October 14th!

To kick things off, we're celebrating on the leaderboard so you can meet the people behind Cloud Run.

A lot of our customers are in India, where WhatsApp often gets a faster reply than email and a phone call still closes more deals than both. With customers in the US and elsewhere, the mix looks different.

We've learned to let each AI employee switch channels: a voice call to qualify, a WhatsApp follow-up with the payment link, an email with the documents.

I'd love real data from other builders. Which channel gets you the best response rate, in which market, and has that changed in the last year?
Support our launch: https://www.producthunt.com/prod...

hey everyone, thank you for launch day. we ended #11 and made the PH newsletter, which i really didn't expect.

a few things shipped since then:

We hear both extremes. Some teams want AI that only drafts, and a human clicks send on everything. Others want full autonomy.
Where we've landed: let the AI act freely on routine work inside clear rules (book the meeting, update the CRM, send the reminder).
Anything outside the rules, like a non-standard discount or a dispute, waits for a named person. The offer stays unchanged while approval is pending.
It's a middle ground, and we keep adjusting where the line sits for each customer.
Where would you draw it? Which actions would you never let an AI take on its own?
SUPPORT OUR LAUNCH!!!- https://www.producthunt.com/prod...

A demo shows what an AI agent can do. A release should prove what it won't do. Before any of our AI employees goes into production, we test three cases:

1. The customer asks for something unapproved, like a non-standard offer. The agent must route it for review and never apply it.

2. A system write fails, for example the CRM update times out. The agent must not report success. It retries safely or escalates.

3. The knowledge base doesn't support an answer. The agent asks a clarifying question or hands off, instead of improvising.

If any of these fail, the release waits. It's slower, but it's why enterprise teams let the agents act at all.

What's on your pre-launch checklist for AI features? What failure taught you to add it?

When UnleashX was handling a few thousand conversations a month, I used to read through a lot of them myself, and I had a pretty good feel for where the agents did well and where they needed work. Once we got into millions, that just wasn t possible anymore, and it completely changed how we think about building.

Speed was the first thing we had to get right, because on a voice call even a one-second pause feels strange, and it has to stay that fast on the busiest Monday morning too. We also stopped trying to review everything and built a way to flag the conversations that actually need a human to look at them, like an unhappy customer or a booking that didn t go through. And we learned that small improvements matter a lot more at this size, since getting even 1% better means thousands more people get what they needed the first time.
Support us here:
https://www.producthunt.com/prod...

The biggest lesson, though, is that customers don t really care how many calls we ve handled. They care that their call went well, and that s what we try to keep in mind every day.

For us, RunEvr is the environment where we actually build RunEvr =)

For me, that s one of the most beautiful parts of having a product - using it while continuing to build and improve it.

We do our everyday calls there, share ideas, chat, store files, do reviews, complete tasks

And all of this is happening inside the same environment we re building to help other teams do exactly that!
Kind of a mirror within a mirror =D

We're debating this internally and would love outside opinions.
One camp says teams want ready-made roles: an AI SDR, a support rep, a collections specialist. You pick one, connect your CRM, set the rules, and it's working in days.
The other camp says every company's process is different, so teams want to define their own roles, tools and approval points from scratch.
We've ended up offering both, but we're still unsure where most teams start. If you were bringing AI into your sales or support team tomorrow, which would you pick, and why?

Hi everyone!

I'm super excited to share our new visual launch editor. It's built to make launching on Product Hunt a lot easier.

We've been building Dots, a macOS app that turns the notch into a customizable workspace.

And honestly, building the app has been the fun part haha.

We've spent countless hours working on features, fixing bugs, and improving the experience. But once we started putting Dots out there, we realized something.

Most calorie apps track you forever. You log, it gives you a number, and you never actually learn anything.

We want Biteme to do the opposite: teach you enough that you can look at a plate and know roughly what's in it, without the app.

Hi everyone! I m Maria, cofounder of Maritime.

We built Maritime because agents need different computers for different jobs: a sandbox to run code, a persistent workspace, or a desktop or phone to use apps with no API.

This is the question we spend the most time on with every new customer at UnleashX.

One sales team wanted their AI employee to offer discounts on its own to close deals faster. Sounded fine until we asked, up to how much? Nobody had an answer. Their human reps had never had a written rule, they just knew from experience.

We ended up setting a limit: up to 40% on its own, anything above that goes to a manager. Deals closed faster, and managers only saw the ones that actually needed them.

Turns out giving AI autonomy forces you to write down rules your team has been following without ever saying them out loud.

Not the dramatic stuff that gets blocked. The quiet things you only notice an hour or a week later

Mine was two agent sessions working in the same repo. One of them went to clean up its own background process by matching the name and killed the other session's run too. No error anywhere. I just found a half finished job later and had to figure out why

My LinkedIn inbox is half spam these days. Pitches right after connecting, fake "quick question" messages, AI written compliments about my profile.

But I still reply to some strangers. Usually when they mention something real I posted or just ask a normal human question without selling right away.

I'm curious where your line is?

What makes you reply to a DM from someone you don't know? And what makes you ignore it instantly?

Our earliest adopters built impressive custom habits around our product's initial limitations, using spreadsheets and manual exports to fill the gaps.

When we finally shipped the native automation to replace those hacks ,engagement actually dropped for a week because their familiar routine were disrupted.

For teams launching core updates: how do you introduce structural product changes without alienating users who rely on the old way?

When users ask for something, it is easy to assume they know the exact solution they need. But building every single requested toggle often leads to a cluttered product rather than better retention.

Lately, I have been trying to dig deeper into the root problem before touching the codebase.

For other founders: what framework do you use to separate genuine product gaps from surface-level feature noise?

Ngl, I'm kinda tired of AI case studies that read like fairy tales. Pilot, rollout, ROI, done.

The numbers tell a different story. MIT's NANDA report (2025) found about 95% of gen AI pilots showed no measurable P&L impact. Gartner expects 40%+ of agentic AI projects to get scrapped by the end of 2027. So where do all the messy middles go?

Every time I look at pre built AI roles for things like customer support or lead qualification the sales page makes it look effortless.

But in reality as soon as you connect your own data and approval rules you realize your internal process is way too messy for a plug and play bot. You end up spending hours re configuring everything anyway which makes me wonder if starting from scratch with modular tools is just better from the start.

For anyone who brought AI into their daily ops did buying a pre packaged role save you time, or did you end up building your own logic?

I spent months on acquisition. Ads, content, outreach, the whole playbook. New signups every week.

Churn ate all of it. I was filling a bucket with a hole in the bottom and calling it growth.

I made something in two days. No plan, no strategy, no thought about whether anyone would use it.

It is still running four years later. Still getting messages from people who use it every week. Still the thing that brings people to my other work.