Video wird geladen...

Video konnte nicht geladen werden

Zur Startseite

We built an ai agent that watches user sessions + emails users when they get stuck - hooked into posthog session replays - every session analyzed w/ gemini - agent detects friction + purchase hesitation - auto-sends the right nudge last week we sent 300 behavior-triggered emails

127,026 Aufrufe • vor 9 Monaten •via X (Twitter)

80 Kommentare

Profilbild von Robert Lukoszko
Robert Lukoszkovor 9 Monaten

Waitlist if you want to try it! I will personally onboard you and support

Profilbild von Elie Steinbock
Elie Steinbockvor 9 Monaten

very cool idea. 1. does it actually work well? (honest answer, please) 2. could be great to catch bugs too. lots of bugs that dont make their way into sentry and similar tools like the user in the email even sent you. i wonder if it could auto notice that error for you

Profilbild von Robert Lukoszko
Robert Lukoszkovor 9 Monaten

1. yes 2. nice thank you! this makes sense

Profilbild von Elie Steinbock
Elie Steinbockvor 9 Monaten

1. excited to try it out if you ever release it as a product!

Profilbild von vas
vasvor 9 Monaten

This is dope

Profilbild von Robert Lukoszko
Robert Lukoszkovor 9 Monaten

Letsgooo

Profilbild von Viz
Vizvor 9 Monaten

As a customer , I wouldn’t want this feature. I don’t like to be tracked.

Profilbild von Robert Lukoszko
Robert Lukoszkovor 9 Monaten

All of this data will be used ONLY be the owner's website. and all PII is included

Profilbild von Sam Grone
Sam Gronevor 9 Monaten

Great idea and extension of existing userneavior analytics platforms like Hotjar/ Posthog. For everyone saying this is creepy, sorry to tell you but most companies record user sessions and run analytics on them (PII is excluded). But things like heatmap tracking, bounce rate, confusion and frustration… recordings are WAY more helpful and accurate than just looking at aggregate metrics or even talking to users. Re: engagement emails - a lot of eComm websites do this already eg “hey you left something in your cart, here’s 10% off”, so that’s standard too. But @Karmedge’s extension to create a merge of these two is quite smart 👏👏

Profilbild von Robert Lukoszko
Robert Lukoszkovor 9 Monaten

ahah! thank you Sam!!

Profilbild von Sam Grone
Sam Gronevor 9 Monaten

For sure! Love the innovation. Maybe worth rephrasing the description to orient towards your customer more, as that may not be clear here, eg: “product managers: we created an automation tool to reduce friction in user onboarding”. Stellar work though and good on you for asking for demoing in public too 🙏

Profilbild von The Canaanite
The Canaanitevor 9 Monaten

Wow! Any chance we can get this open-source?

Profilbild von Robert Lukoszko
Robert Lukoszkovor 9 Monaten

Interesting, I will think about how to implement it! dm'ed you!

Profilbild von Avi Peltz
Avi Peltzvor 9 Monaten

This is sick!

Profilbild von Robert Lukoszko
Robert Lukoszkovor 9 Monaten

lfg

Profilbild von Daniil
Daniilvor 9 Monaten

Thanks man, add to backlog ✅

Profilbild von Robert Lukoszko
Robert Lukoszkovor 9 Monaten

nice

Profilbild von Abhinav Singh
Abhinav Singhvor 9 Monaten

Dope shit my guy dope shit.

Profilbild von Robert Lukoszko
Robert Lukoszkovor 9 Monaten

thank bro!

Profilbild von Saleh Hindi
Saleh Hindivor 9 Monaten

Looks super useful, love to see it! Creative use of gemini's video processing

Profilbild von Robert Lukoszko
Robert Lukoszkovor 9 Monaten

Thanks! yeah, video is interesting

Profilbild von Piotr Gerke
Piotr Gerkevor 9 Monaten

Why email? Make it take action on user’s behalf and purchase my product lmao

Profilbild von Robert Lukoszko
Robert Lukoszkovor 9 Monaten

hahahah, for suree

Profilbild von Fritz
Fritzvor 9 Monaten

Fantastic work mate!!

Profilbild von Robert Lukoszko
Robert Lukoszkovor 9 Monaten

thanks Fritz

Profilbild von Tasfia ⋆.˚⋆
Tasfia ⋆.˚⋆vor 9 Monaten

This makes a lot of sense. Curious are the nudges 100% automated or can a human step in before sending?

Profilbild von Robert Lukoszko
Robert Lukoszkovor 9 Monaten

yes, we have will have a review/edit/approve option

Profilbild von Daniel
Danielvor 9 Monaten

Holy shit thats crazy

Profilbild von Robert Lukoszko
Robert Lukoszkovor 9 Monaten

thank bro

Profilbild von Vlad 🍩
Vlad 🍩vor 9 Monaten

how many replies did you get?

Profilbild von Robert Lukoszko
Robert Lukoszkovor 9 Monaten

so far around 10

Profilbild von ian
ianvor 9 Monaten

ngl, this creeps me out as a consumer. way too big brother-esque.

Profilbild von Robert Lukoszko
Robert Lukoszkovor 9 Monaten

Companies gotta get those customer back, no matter what method

Profilbild von ian
ianvor 9 Monaten

certainly don't disagree there, but i think if you can leverage that data without sending them a clearly overly-targeted email about what caused them to drop off, you'll have much more success and avoid looking like you're explicitly tracking them

Profilbild von ian
ianvor 9 Monaten

and as a developer, absolutely love the thinking, just take issue with how you use the info

Profilbild von Robert Lukoszko
Robert Lukoszkovor 9 Monaten

thanks man

Profilbild von dolev
dolevvor 9 Monaten

very smart

Profilbild von Robert Lukoszko
Robert Lukoszkovor 9 Monaten

grinding with team out there 24/7

Profilbild von Don Park
Don Parkvor 9 Monaten

good idea. creepy too, unfortunately.

Profilbild von Robert Lukoszko
Robert Lukoszkovor 9 Monaten

what makes it creepy? i dont think it is

Profilbild von Don Park
Don Parkvor 9 Monaten

mechanism itself is fine and rather innovative. lack of user approval made me assume this feature would be opt-out at best. that's creepy to me. responses will vary, of course. take it as a data point, not a general verdict.

Profilbild von Jukka
Jukkavor 9 Monaten

Very cool but also slightly creepy at the same time. How do you spot the user’s intent to nail the emails?

Profilbild von Robert Lukoszko
Robert Lukoszkovor 9 Monaten

what makes it creepy? I feel it is normal

Profilbild von Jukka
Jukkavor 9 Monaten

At first it was the feeling of being spied on. Now that I think of it, what you have is about using tracking data differently than most do. As a part of the product, in an active way. Hotjar and similar are already tracking cursor movements etc. but how the data is used is passive, and definitely does not help the user immediately.

Profilbild von Robert Lukoszko
Robert Lukoszkovor 9 Monaten

Yes thats what I am thinking as well. this data is being owned by website. Wonder what do you think the most active ways to use it are? maybe we can continue convo in dm's

Profilbild von Jukka
Jukkavor 9 Monaten

Sure let’s do that

Profilbild von owen
owenvor 9 Monaten

this is sick

Profilbild von Robert Lukoszko
Robert Lukoszkovor 9 Monaten

hell yeah

Profilbild von Julia Seregina
Julia Sereginavor 9 Monaten

Great one for product teams!

Profilbild von Robert Lukoszko
Robert Lukoszkovor 9 Monaten

@julia_in_purple we will kill amplitude?

Profilbild von Julia Seregina
Julia Sereginavor 9 Monaten

Or product analysts? 😜

Profilbild von Jelmer Borst
Jelmer Borstvor 9 Monaten

Love it! Great idea, all the best

Profilbild von Robert Lukoszko
Robert Lukoszkovor 9 Monaten

thank you!

Profilbild von Ron
Ronvor 9 Monaten

Great idea! I thiiink for the first N emails I'd want to personally edit+send the message for stylistic reasons, make sure I'm around for a live chat if the customer wants it, etc.

Profilbild von Robert Lukoszko
Robert Lukoszkovor 9 Monaten

absolutely! 100% agree. approve/edit/reject

Profilbild von Pochi
Pochivor 9 Monaten

interesting use case. curious how you think about keeping the nudges feeling helpful vs intrusive and what the false positive rate on "purchase hesitation" looks like right now 🤔

Profilbild von Robert Lukoszko
Robert Lukoszkovor 9 Monaten

you can program it with prompts!

Profilbild von Ray
Rayvor 9 Monaten

Interesting

Profilbild von Robert Lukoszko
Robert Lukoszkovor 9 Monaten

interesting

Profilbild von Ricky Funk
Ricky Funkvor 9 Monaten

Developers: "This is great" Customers: "Remove this shit RIGHT NOW" Listen to your customers. Get feedback. Leave the Bubble.

Profilbild von Robert Lukoszko
Robert Lukoszkovor 9 Monaten

@rickyfunk Getting feedback as we speak!!!

Profilbild von m ø w
m ø wvor 9 Monaten

If there’s friction, is not necessary to address the friction instead of survalance? I mean I might be wrong, and often am but I’m interested in how other people set up their systems architecture, especially when considering second order consequences.

Profilbild von Robert Lukoszko
Robert Lukoszkovor 9 Monaten

sorry didn't get this. but what do you mean by fritctin?

Profilbild von Mick
Mickvor 9 Monaten

this is really cool. how do you do the auto zoom when doing looms for demos?

Profilbild von Robert Lukoszko
Robert Lukoszkovor 9 Monaten

screen studio!

Profilbild von Mick
Mickvor 9 Monaten

champion! thanks bro :)

Profilbild von darren
darrenvor 9 Monaten

such a cool idea

Profilbild von Robert Lukoszko
Robert Lukoszkovor 9 Monaten

Thanks man! Appriciate

Profilbild von Skyler Ji
Skyler Jivor 9 Monaten

seems super exciting!

Profilbild von Robert Lukoszko
Robert Lukoszkovor 9 Monaten

thanks! Doing my best work out there

Profilbild von Brad | AdaptiveX Voice Agents
Brad | AdaptiveX Voice Agentsvor 9 Monaten

Fantastic! I built something similar as a poc earlier this year and have it in my pipeline. The difference is instead of email it's a voice agent that nudges them to have a quick chat, exactly like a team member in a physical store. I love personalised nudges, well done Robert!

Profilbild von Robert Lukoszko
Robert Lukoszkovor 9 Monaten

thanks!!

Profilbild von Barron
Barronvor 9 Monaten

Love this

Profilbild von Robert Lukoszko
Robert Lukoszkovor 9 Monaten

Letsgooo

Profilbild von Anayat
Anayatvor 9 Monaten

Behavior agents watching PostHog replays is such a sweet spot for LLMs, next level is closing the loop in-product too so the same agent can trigger UI nudges or micro-surveys before it ever needs to send an email

Profilbild von finity
finityvor 9 Monaten

What about opening a chat popup that says need help and giving summary of what they did wrong + chatting to it?

Profilbild von Robert Lukoszko
Robert Lukoszkovor 9 Monaten

Smart! Noted

Profilbild von finity
finityvor 9 Monaten

Would love to be a beta user! And give feedback

Profilbild von logleb
loglebvor 9 Monaten

this is sick

Profilbild von Robert Lukoszko
Robert Lukoszkovor 9 Monaten

thanks! hard work!

Ähnliche Videos

This is insane. An AI agent can run every boring job in outbound. We spent the last 6 months building ours. Here are my 8 favorite agents to build: Replies get sorted before we open the inbox. Campaigns go live from one command. Weak inboxes pull themselves out before they hurt a domain. Here are the agents behind it: 1. Reply Agent Reads every reply and drafts the response. A human reviews, edits, and sends. 2. Mailbox Health Agent Watches inbox and domain health. It predicts when you need new mailboxes, then buys and warms them. 3. Campaign Optimizer Agent Checks every live campaign every 6 hours. If 500+ leads were emailed and replies are under 4%, it tests new copy, replaces inboxes replying under 1%, and shifts sending to better hours. 4. Lead Qualification Agent Scores each lead by company size, industry, and tech stack. It enriches the record, updates your CRM, and only loads qualified leads into campaigns. High-priority prospect? You get a Slack ping. 5. Meeting Booking Agent Finds meeting requests inside replies. It books the slot, writes prep notes using the lead's background, sends reminders, and logs the outcome. 6. Pipeline Progression Agent Tracks opens, clicks, and website visits. It moves the CRM stage, triggers the next sequence, and creates a task when a lead shows real intent. 7. Copywriting Agent Writes cold emails and follow-ups in the campaign's voice. 8. Analytics Agent Watches campaign metrics in real time and explains what to fix in plain language. We built ours with custom code. Smartlead's SmartAgents let you build agents like these from a plain-English prompt, inside the platform where your campaigns already run. If you repeat an outbound task more than twice a week, that is an agent you have not built yet. Which one would you build first?

Hosun Chung

156,625 Aufrufe • vor 2 Monaten

We use OpenClaws to do all of our work at Every 📧. We have 25 full-time employees, so we’re one of the few companies in the world that has seen how work changes when everyone has their own personal agent in the company Slack. I chatted with Every 📧 COO Brandon (Brandon Gell) and Every 📧 head of platform Willie (Willie) to share what we’ve learned. We get into: - Why agents become mirrors of their owners, and how that influences how other people on the team interact with them - How a parallel AI org chart forms on its own. People have stopped tagging me on Slack with questions about Proof, the document editor I vibe coded, because they knew my agent R2-C2 can step in - The etiquette for human-agent collaboration is being invented in real time. Brandon's rule is that if there's an established process or documented answer, always ask the agent, not their human - Why everyone is a manager now, and why even experienced managers carry limiting beliefs about what their agents can do - This is a must-watch for anyone trying to understand how AI workers change daily operations, not just in theory, but inside a company that’s half-agent Watch below! Timestamps Introduction: How Brandon built Zosia, an AI agent to run his household: Brandon’s “aha” moment: What happened when everyone on the team got their own agent: How agents take on their owners' personalities, and why that matters inside an org: Why it’s important for agents to work in public: What we’re still figuring out when it comes to agent behavior, including memory gaps, group chat etiquette, and the "ant death spiral" problem: How we built Plus One, our hosted OpenClaw product: The cultural shift required to make agents work at scale:

Dan Shipper 📧

67,958 Aufrufe • vor 5 Monaten

There are 8 billion people on earth. Soon there'll be 100 billion AI agents. Every one of them needs email. Six weeks ago I said the next wave of teams would run email through an agent instead of a dashboard. Today it ships. Nitrosend☄️ is launching Agentic Email Marketing: the email layer for the agent economy. What agents can do on Nitrosend right now: Sign themselves up. Point any agent at and it creates the account, connects your domain, sorts billing and sends its first email. No API key. No dashboard. No human required. Shipped, and users agents signing up with it daily. Get their own inboxes (beta, by request). Real addresses on the domain you own. Your agents receive, and send 1-1 email conversations with customers. A reply lands at 3am, your agent answers it. Anything that needs a human gets escalated to you. Ask us and we'll flick yours on. Next: Agentic Outreach (coming soon). Your agent studies your best customers, finds more like them, writes like a person, sends in sequence and works the replies. Then: set a goal and walk away. Goal-based agentic marketing is in development. "20% more activations this quarter" and Nitrosend plans, sends, measures and improves every week. Why we built this: Gmail is agent hostile and expensive per seat. Legacy email platforms assume a human sitting in a dashboard. agents needed an email layer of their own. They're already better at it than we are. They read everything, never miss a follow-up, and write personally at any scale. *94%* of actions on Nitrosend already happen inside an agent (Claude, Codex, ChatGPT, Cursor), not in our UI. Humans approve. Agents operate. This is our third email company. Six billion emails across the first two. We've been burned by every ugly part of email already, which is why the approval gates are built in exactly where you want them. Watch the launch, then send your agent to work: send it.

George Hartley ☄️

945,911 Aufrufe • vor 2 Monaten