Pebbles Ai
AI sales platform for modern B2B teams
701 followers
AI sales platform for modern B2B teams
701 followers
More pipeline, faster deals, fatter margins. Pebbles Ai is a Go-to-Market Operating System™ that changes how commercial teams work. One place replacing 10+ tools powered by neurosymbolic AI. Whether you're two co-founders, a 10-person startup, or a 100 strong company, Pebbles Ai truly moves the commercial needle without breaking the bank. Generate strategies grounded in science. Create content that earns attention. Launch campaigns infused with persuasion. Build sales assets that close deals.
AISA AI Skills Test
the unified workspace angle is smart — the real pain in B2B outreach isn't any single tool being bad, it's the context-switching between 6 different ones. curious how you handle the AI personalisation at scale without it starting to feel templated? that's usually where these systems start losing reply rates
Pebbles Ai
@ozandag VERY sharp. You've put your finger on the exact failure point.
The way I'd frame it: there are three types of AI tools on the market right now.
Wrappers. A thin UI over a raw model. You're basically prompting ChatGPT with extra steps. Fast, but zero real context
AI-templated wrappers. Smarter, but still template-first. They spin variations on a fixed skeleton and drop in variables. This is the category you're describing, and you're right, at scale it starts to feel like a mad-lib, because underneath it's the same message wearing different hats
Neurosymbolic AI. This is us, and it's a different mechanism entirely. It doesn't fill a template. It reasons each message from the actual person: who they are, what they care about, what the strategy layer knows about their segment, and the rules for their persona. The structure itself changes, not just the variables
That's the whole difference. A templated system personalises the blanks. A reasoning system composes the message. One drifts toward sameness at volume because the skeleton is fixed.
The other holds quality at scale because there is no skeleton, every message is reasoned from scratch against real context.
So the reason reply rates hold is architectural, not a clever prompt. The thing that makes outreach feel templated is precisely the thing we removed.
It's a totally different architecturally solution.
Short version: wrappers give you speed, templated wrappers give you speed with a costume, and neurosymbolic actually reasons per person, and doesn't break.
That last one is the only category that survives scale.
Congratulations on a launch! I have used a similar project before, but it had very limited features. You look very solid and look exactly like what I need.
P.S. Dashboard looks nice, but some extra "llm" prompts showed me a bit confused. Like asking for ICP and then fetching from the data we already gave
Pebbles Ai
@spiri7 Thank you, that really means a lot. And congrats right back for being an innovator-minded person that likes to try novel things.
Hearing it looks like exactly what you need, especially after a thinner tool let you down before, is the best thing I could read this afternoon.
Now, your P.S. That's the most valuable part of your whole comment, so thank you for flagging it. Honest UX feedback like this is how the product actually gets sharper.
You're right that it can feel like doubling up. There is a reason behind it: the ICP step is about sharpening who you target, which is subtly different from the raw company data you gave us. But if it read as "didn't you just ask me this," that's a gap on our side to close, not yours to decode.
So here's what I'd love to do. Let me give you a 30 min walkthrough. I'll show you the tips and tricks, why a few steps exist, and how to get the system really flying for your company.
Would you like that?
For small B2B teams, the sales-agent win is usually handoff quality rather than full autonomy: what changed, what evidence supports it, what needs human judgment, and what should never be sent without approval.
Pebbles Ai
Great question Patrick!!! Love when technologists go deep.
For small B2B teams, full autonomy is mostly a pitch-deck fantasy today, and the reason is arithmetic, not opinion.
Agent errors compound multiplicatively across steps. At 95% reliability per step, a 10-step task lands around 60% overall (0.95^10). Even at an optimistic 99% per step, a 20-step chain only succeeds ~82% of the time.
Empirically it's worse: in Carnegie Mellon's TheAgentCompany benchmark, even the strongest model completed only ~30% of realistic, multi-step office tasks end-to-end (earlier runs landed at 24%).
And the market is pricing it in. Gartner expects over 40% of agentic AI projects to be canceled by the end of 2027, citing unclear value and inadequate risk controls, and notes today's models can't reliably follow nuanced instructions over long horizons.
People are frustrated, and frankly, when I talk to customers, angry!
So the winning pattern isn't more autonomy. It's selective autonomy: aggressive automation on low-stakes steps, hard human gates on high-stakes ones, with escalation paths that cap the blast radius. Which is precisely the handoff quality you're describing.
You framed these four around the sales agent, but they're really the design spec for the whole platform, strategy, marketing and sales alike. So here's how each one works at that level:
What changed?
Whatever the surface, a market shift the strategy core detects, a competitor move, a new buying signal, an inbound reply, the platform isolates the delta and leads with it. You get the one thing that moved and why it matters to YOU, not a raw feed you have to reverse-engineer. Same behaviour whether it's your positioning or your pipeline.
What evidence supports it?
Every recommendation carries its reasoning, a strategic call, a beachhead segment, a line of outbound copy. This is where neurosymbolic earns its place over a raw LLM: the symbolic layer keeps each inference traceable back to the rule or data point that produced it, so a conclusion arrives with its "why" attached rather than as an unexplained assertion from a black box. Inspectable reasoning like that is what the research points to as the precondition for trustworthy human-in-the-loop oversight.
What needs human judgment?
Whether it's a strategy decision, a targeting call, or an outbound message, when confidence is low (we have "judges") or the call is genuinely ambiguous the platform surfaces that explicitly and routes it to you, instead of papering over uncertainty with a confident-sounding guess. Overconfidence in failure is one of the documented ways agents break, so we treat "I'm not sure, your call" as a feature, not a bug.
What should never be sent without approval?
Anything irreversible or externally facing, a published campaign, an outbound send, an investor or sales asset going out the door, is gated by default. The platform can research, reason, write, sequence and prepare the whole thing, but the irreversible action stays with a human until you explicitly choose to loosen it.
Notice the common thread: every one of those four depends on reasoning you can actually trust, not a confident guess. That trust is the exact thing the arithmetic up top says nobody has yet, it's why 40% of these projects get canceled. So the real question is whether our engine is any different. Here are the numbers.
How does our neurosymbolic actually perform?
A base LLM is inherently a chatbot system. It predicts the most probable next word, brilliantly, but that is not enough for complex domains such as B2B GTM, Corporate Law, and Human Resources. Nothing in it stops to ask "is this true, and does it obey the domain rules".
Neurosymbolic AI is 3-step systems working together, with a check between them:
The neural half reads language and context, the way any strong LLM does
The symbolic half applies explicit logic, rules and a structured knowledge base
A verification step sits between that and the output, catching anything that breaks
So you get the fluency of an LLM with a reasoning and fact checking layer bolted underneath. Outputs are accurate, reproducible and explainable (even auditable) rather than a confident black box.
It is also rare: academic interest went from 112 papers in 2015 and 2016 to over 9,000 in 2025 and 2026 [Google Scholar], yet real production systems are almost nowhere, because building one needs machine learning, formal logic, knowledge engineering and domain science in the same room at once.
Let's look at some numbers. First, let's compare Claude Opus on the Max tier with a single instruction against the full Pebbles pipeline:
Accuracy: 33% vs 87%
Precision: 57% vs 91%
Sales Efficacy, MQL to SQL: 15% vs 85%
And at the architecture level:
82% lower error rate than LoRA fine tuning, because the architecture is structurally accurate rather than nudged
3x better gross margin than wrappers, because the reasoning is not rederived from scratch on every call
~2% hallucination on rule bound queries, versus 31.4% across real world use HalluScore benchmark
But don't take my word for it: Claude Opus sits around 33% factual hallucination on the public HalluScore benchmark, while generally neurosymbolic methods approach ~100% accuracy on rule based tasks [arXiv 2502.01657]. We are closing up at 85-98%.
Cost is where it gets almost silly. To rebuild one reply with a raw model:
Around 12 prompts per reply, each re-sending 25,000 to 35,000 tokens of context
Roughly $5 to $7 per usable output, and that's not even top 1% reasoning
About $5,000 to $7,000 a month at that volume, before 500 hours of human prompting
THIS MEANS WE CAN GIVE MUCH MORE AI ALLOWANCE TO OUR USERS!!! 🫶🏻❤️📈
Finally, the neurosymbolic reasoning is what lets it carry complex, multi-faceted, and cross-functional B2B work all the way through. These are examples you can build and execute on, which is impossible with base-models or tools with wrappers:
A full go to market strategy, grounded in your ICP, positioning and live market signals, then turned into the campaigns that run it
A beachhead strategy, picking the wedge segment worth attacking first, sizing it, and sequencing the entry instead of guessing
Industry trend analysis read across macro, meso and micro signals, so you see the shift before it hits your pipeline
Investor decks and enterprise sales assets, two pagers, RFPs and proposals that hold up when a sharp reader pushes on them
An omnichannel outbound engine, from fresh leads to reasoned email and LinkedIn sequences to replies captured and qualified in Smartbox, built and run end to end
And this is only the current stage. We are making the first steps toward a true Jarvis for go to market: a system that can safely, securely and reliably run the work fully autonomously, with no human in the loop.
I hope that makes sense.
Finally tried Pebbles Ai and the campaign planning flow is way smoother than I expected. Having strategy and outreach in one workspace cut my tab switching way down.
Pebbles Ai
@nhanife56159 This made my day, thank you! The tab-switching was the exact thing that drove me mad as a founder, so hearing that the strategy-to-outreach flow cut it down for you is the best feedback we could ask for.
That "one workspace" feeling is the whole reason we built it, so it means a lot that it is received well. If you hit anything confusing or have a wishlist as you go deeper, I'm all ears. We're shaping it around real users and we ship fast.
The positioning keeps referencing enterprise teams and McKinsey-level GTM strategy, but the problem you're describing first customers, fragmented tools, no dedicated sales team is most acute for founders at the zero to one stage who can't afford £1499 a month. Curious whether there's a solo founder or early stage tier planned, because that's the exact audience most burned by the tool fragmentation problem you're solving.
Pebbles Ai
@jasnoor_singh_oberoi There's good news. The £1,499 isn't the entry point. That's the Organisation tier, built for a 10-seat commercial team, often for lower-mid to mid-market organisations.
The way in is Pro, at £49 a month. Even better, the Team subscription at £349 a month has the most value. The most bang for your buck.
We wanted to make the value stupidly high. Like a NO-BRAINER. Something that would absolutely flabbergast new users. How is it possible to add so much value at such a low price.
(Spoiler alert: it's possible because our system is built on neurosymbolic AI)
So the exact person you're describing, the solo founder buried in fragmented tools with no sales team, isn't an afterthought for us. They're arguably who we built this for first. Don't forget, we were also in that position.
In our experience, solving for smaller teams has proven to be much more difficult than for larger teams, ironically. So we did it the hard way first. Naturally haha.
On the positioning itself, you're right. When we say "enterprise-grade" or "McKinsey-level," we mean the quality of the reasoning and judgment, not the size of the customer or the price of the plan. We should be more clear.
Our mission is David versus Goliath. Give a two-person startup the same GTM firepower a big company pays top-tier consultants and 30 premium tools for, at a price a founder can actually afford.
In short, you only scale up when your team does.
Best way to judge it is to run it yourself against the mechanic and overwhelming tool fragmentation problem you described.
Start free, no credit card needed. I'd recommend trialing the Team subscription.
You won't regret it. 😊
Pebbles Ai
@akshitha_kalwakunta Immediately. The moment you update your positioning or brand voice in the Company Library, every core picks it up.
The next outreach and every Smartbox reply reflect the new guidelines straight away, no waiting for days. One edit and the whole system moves with you.
Pebbles Ai
Thank you for stopping by our Pebbles Ai page. This is our story.
Three years ago, we set out to build something the market simply didn't have. We did it the hard way.
We built the brain first, the neurosymbolic AI, then wrapped the operating system and the workflows around it. Everyone else builds the car and forgets the engine. We built the engine, then realised we should probably add doors.
We didn't wrap a general-purpose LLM in a logo and call it a Series A. We refused. To make things more difficult, we built it across two cities: London (UK) and Lviv (Ukraine).
If you've been following what's happening in Ukraine, you know what that means. Our team has built the smartest GTM solution on earth through air raid sirens, blackouts, and nights filled with whistling ballistic missiles.
They never stopped shipping.
Honestly, the team jokes that the war wasn't even the hard part. Building something this new, this wide, and this deep all at once, an entire operating system grounded in empirical science, from scratch, that was the truly terrifying bit. The missiles were just the background music.
So why would you put yourself through all of that?
In hindsight, therapy would have been cheaper. But less scalable.
Because the commercial game is rigged. Enterprise companies hoard the best talent out of universities, pay obscene salaries, and keep the most advanced tools for themselves.
The rest of us? Mere peasants in a story of oligopoly kings.
The result is the erosion of small and mid-sized businesses. And just like the erosion of the middle class, that's bad for capitalism, bad for the economy, and frankly bad for democracy.
So we wanted to level the playing field. To give David the slingshot he deserved, our technology, so he can take Goliath down a peg and steal real market share without breaking the bank.
We're not throwing stones. We're slinging lethal pebbles at Mach 3 speed.
That's the whole point. Enterprise-grade go-to-market firepower, in the hands of the people with big ambitions and great products.
So how do I know you won't just burn my money?
We're not the startup that raised millions and blew it on ping-pong tables and Super Bowl ads.
We raised millions and spent it on the least glamorous thing imaginable: making it actually work. 87% went straight into the technology. The team took pay cuts. The founding team forfeited salary for three years. Our accountant thinks we're a charity.
Because we'd rather give the world something that actually works. This was never going to be easy, and we needed every dollar in the tech.
So what did you actually build?
The world's first Go-To-Market Operating System (GTMOS™), powered by 8 neurosymbolic AI cores, where management, marketing, and sales all work together in one workspace.
We went wide: one platform that replaces 10+ tools. And we went deep: real reasoning under every feature. One stack to replace them all. Sauron would be jealous of the licensing savings.
Every capability is custom-built for precision, accuracy, and commercial efficacy. And like real departments, they talk to each other.
So you can replace your whole stack with one GTMOS, pay less than you do today, and get an intelligence that understands your company better over time.
One last thing. Your assistant isn't built to agree with you. It's built to make you succeed. It will push back. It won't stroke your ego, because it can't stroke your ego and make you win at the same time.
It only cares about your success, not your feelings. Refreshing, we know. These are autonomous, top 1% domain-expert thinkers, built to guide you through the maze of go-to-market.
So who's actually behind this?
I'm endlessly proud of the people who built this, many of them while their country is at war:
Dmytro Antoniuk, our CAIO @dima_antoniuk
Maksym Kuzmovych, Head of Engineering @kuzmovych
Priyanka Mandal, Head of Community @priyankamandal
And thank you to every early adopter showing interest in Pebbles Ai. I'm truly humbled, and I appreciate your time and intellectual curiosity.
Go sign up, it's free. Built for people with big ambitions and small teams. Emphasis on small. Push the system hard. Tell us what you love, tell us what you want more of, and help us shape it around you.
💛 If you're an entrepreneur: just follow and message me. I will personally give you a 30-minute demo and show you how Pebbles genuinely improves your professional life, strengthens your team, and transforms your company.
Whether you are an early-stage startup of two co-founders or a mid-market organisation with 1,000+ employees, you will see how science and technology applied to GTM can move the commercial needle without breaking the bank.
Or, alternatively, if you would rather not talk to anyone:
Head to Pebbles Ai and discover it without any cost ("start free" button)
No credit card required. A generous AI allowance. Up to 3 seats included
And 3,000 fresh leads to get you started!!!!!!!!!
🚀 PRODUCT HUNT EXCLUSIVE. 24 HOURS ONLY.
Pebbles Ai
For investors (VCs/Angels) we are incidentally also raising our first institutional round (Seed). Get in touch!