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
Thought I was fixing an API issue… turned out the real problem was in the database. 😅
Checked the API logic, traced the request, and kept debugging only to find out the data itself was the problem.
Sometimes the hardest part isn't fixing the bug. It's finding where it actually lives.
Ever had a bug send you in the completely wrong direction? 👀
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
Related posts
Most AI apps work like this: the model decides, the app saves, and the user finds out later that the numbers are wrong. 😬
For Notarik, I set one rule. The AI can only propose. Nothing gets saved until the user confirms. ✅
🧾 What it does
Take a photo of a receipt, or type "lunch 45k at the warung". You get a categorized expense, a monthly recap, and a chat assistant you can ask things like "where did my money go this month?":
🔒 The parts I care about most
• Receipt totals are checked by plain code, not by the model. If the math doesn't add up, it gets flagged.
• Money is stored as whole numbers (IDR), never decimals, so there are no rounding bugs.
• Every user's data is locked at the database level with Postgres row-level security. Even if the app has a bug, it can't show someone else's records.
• Text on a receipt is treated as untrusted. If a receipt says "ignore your instructions", it's just text.
• The agents have tests, so when I change a prompt I can see right away if it got worse.
🛠 Built with Next.js, React Native, Python (FastAPI), LangGraph and PostgreSQL. One person, web and mobile, from design to API.
🤝 If you're building something where an AI touches real money, records or user data, this is the kind of work I like: agents that are useful and also safe to put in front of users.
👉 Try it live: https://notarik-app.nashir.tech
Want to talk about yours? Send me a message.
Smart call keeping the confirmation.
I gave a dental clinic a receptionist that never sleeps. 🦷
Today I finished building a WhatsApp AI Receptionist — and tested it live, end to end.
A patient messages the clinic on WhatsApp. Within seconds they get a polite, human-sounding reply with the clinic's REAL fees, timings and doctors — because the AI answers only from the clinic's own database. It never invents a price. If it doesn't know, it says it will confirm with the owner.
The honest part? Halfway through testing, it broke. Spectacularly. 😅
WhatsApp sends invisible "delivered" signals — and my system was answering those signals too. 192 executions later, I found the cause hiding in one execution log: no message, just a delivery status. One guard node later, the loop was dead and the replies were clean.
That debugging session taught me more than any tutorial:
→ Always guard your trigger. Not every webhook event is a message.
→ Read the raw execution JSON before guessing. The answer is always in there.
Now I'm setting this up for clinics and local businesses that live on WhatsApp — instant replies, appointments captured, every lead saved, even at 2 AM.
If your business answers the same 10 questions on WhatsApp every day — what would you automate first?
