The network for creativity
Join 1.25M professional creatives like you
Connect with clients, get discovered, and run your business 100% commission-free
Creatives on Contra have earned over $150M and we are just getting started
Back to feedPost
A production multi-tenant AI customer-engagement platform I designed and built end to end. A chatbot answers customer questions 24/7 from each business's own knowledge base (RAG), captures leads, sends instant Telegram alerts, and hands off to a human in real time. Includes a voice assistant and a per-client analytics dashboard.
Self-hosted — each business fully owns its instance, code, and data, with no SaaS lock-in.
Built solo: architecture, backend, AI integration, infrastructure, deployment, and security.
Stack: Next.js, Directus, PostgreSQL (pgvector), Redis, n8n, Twilio, OpenAI / Claude / Gemini, Docker.
The network for creativity
Join 1.25M professional creatives like you
Connect with clients, get discovered, and run your business 100% commission-free
Creatives on Contra have earned over $150M and we are just getting started
Challenges
View all
Related posts
: Hi everyone!
AI Lead Qualifier — chat agent that qualifies leads before your manager picks up the phone.
THE PROBLEM
78% of buyers choose the vendor who answers first. Most inbound leads wait hours for a human.
THE SOLUTION
An AI agent answering every lead within seconds, 24/7: short dialog → structured data (need, budget, timeline, contact) → A/B/C grade → instant Telegram alert with a full lead card. Everything logged to Google Sheets.
BUSINESS IMPACT
• Response in seconds, not hours
• 50–70% of routine questions off your sales team
• Structured lead card for every inquiry
• Self-hosted — no monthly SaaS fees
PROVEN IN A LIVE TEST
From "can you make a website for my store?" the agent extracted need, timeline, budget, name and phone in 2 messages, scored the lead class A, and alerted the manager — automatically.
Built on n8n + LLM + Telegram. Details in my profile — DM me.
I’ve been building SabiFlow for a while, and honestly, the most interesting part of the project isn’t the code.
It's the problem.
Because it's personal.
I've experienced that thing where money comes in and somehow, without you really noticing, it starts disappearing.
Not because you don’t earn enough.
Not necessarily because you're irresponsible either.
Sometimes money simply has no job when it arrives.
And as an engineer, that got me thinking:
What if the problem isn't budgeting? What if the problem is that we’re asking people to make too many good decisions at the exact moment they have the most temptation to make bad ones?
That question became the foundation for SabiFlow.
Instead of telling someone, "You should save 20% of your income," I started thinking about what would happen if the system simply helped assign every inflow a purpose the moment it arrived.
That led me down a rabbit hole.
Funnels.
Automated distribution.
Wallet infrastructure.
Virtual accounts.
User behaviour.
Transaction flows.
KYC.
Compliance.
Even the psychology behind notifications.
And this is probably my favourite part of being both an engineer and a founder.
I don’t just ask:
"How do I build this feature?"
I ask:
"Why does this problem exist, and what kind of system could make dealing with it easier?"
Then the engineer in me comes along and asks:
"Okay… but how do we actually make this work reliably?" 😂
That tension between the founder thinking about the problem and the CTO thinking about the system is probably what I enjoy most about building SabiFlow.
I’m still figuring a lot of it out.
But I'm curious:
What’s a problem you’ve experienced personally that eventually made you want to build something around it?
The framing around “why does this problem exist?” is exactly the kind of product thinking that keeps a financial tool from becoming another dashboard. SabiFlow sounds strongest where the behavioral insight meets the practical system design—especially around notifications and reliable follow-through.
I built ReviewIQ because technical interviews don't always test whether you can actually review code.
ReviewIQ is an AI-powered code review interview trainer built for software engineers.
You pick a role, language, and seniority, then get a realistic PR diff with bugs intentionally planted in it.
You write your review.
Then the system grades it against the actual bugs, shows what you caught, what you missed, and gives you feedback on how a stronger reviewer would approach it.
The interesting part was building the grading system so it isn't just "AI thinks your answer is good." The bugs have a known ground truth, so the review can be evaluated against something concrete.
Built with Next.js, Supabase, PostgreSQL, OpenAI, and Lemon Squeezy.
I'm interested in building more products where AI isn't just a chat box, but an actual part of the product experience.
The known-ground-truth approach is a great product decision—it makes the feedback feel earned rather than like an opaque AI verdict. I also like that the flow tests the actual review skill instead of rewarding pattern-matching in interviews.