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
Static portfolios showcase skills, but smart portfolios prove them in real-time. ⚡
Here is a quick look at the custom AI Assistant built directly into my portfolio web application!
Key Features & Stack:
• Real-time context retrieval for interactive Q&A
• Ultra-fast AI response latency powered by Groq API & LLM integration
• Seamless Next.js floating UI with dark mode hierarchy
Integrating custom AI agents directly into web interfaces increases visitor engagement and automates lead generation effortlessly.
How are you implementing AI capabilities into modern SaaS & web apps? Let's connect! 👇
This is the right direction — a portfolio that answers back proves more than screenshots ever will. The real-time context retrieval is the key part: it only works if the assistant answers from your actual work, not generic model knowledge. In my own agentic builds, that...
Thanks for the thoughtful comment, Olanrewaju! Spot on grounding is exactly what separates a useful agent from a novelty chatbot.
To answer your question: when a visitor asks something outside my portfolio scope, the assistant strictly responds with: "I'm only trained to talk...
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
🚨 A dental clinic was losing patients every single day.
Not because of bad service. Not because of high prices.
Because nobody was replying to messages.
Patients messaged on WhatsApp, Messenger, SMS… And waited. And waited. And left.
The clinic owner contacted me with one simple ask:
“I can’t hire someone to reply 24/7. Can you help?”
So, I built them an AI receptionist that never sleeps. 👇
Here’s exactly what it does:
✅ Omnichannel Support: Instantly replies to patients on WhatsApp, Facebook Messenger & SMS — all from one system.
✅ Human-like AI: Google Gemini AI understands the message, detects intent, and replies naturally.
✅ Smart FAQ Handler: Answers questions about services, pricing & clinic hours by pulling live data from Google Sheets.
✅ Automated Booking: Checks availability, blocks double-bookings, and confirms only after saving to the sheet.
✅ Instant Alerts: Notifies the clinic owner instantly via Gmail + Telegram on every new booking.
✅ Smart Reminders: Sends automated email reminders — 2 hours and 1 hour before the appointment.
✅ No-Show Recovery: Detects missed appointments and automatically sends a polite rescheduling email after 12 hours.
The Tech Stack 🛠️:
⚙️ Make.com — Automation backbone
🧠 Google Gemini 2.0 Flash — AI brain
📊 Google Sheets — Live clinic database
💬 Green API — WhatsApp integration
📱 Facebook Messenger API & Twilio — Chat & SMS
✉️ Brevo — Email reminders
📢 Telegram Bot — Owner notifications
💾 Make.com DataStore — Conversation memory per patient
The Result? 📈
Every patient gets an instant, intelligent reply — 24/7. No missed bookings. No forgotten appointments. No manual follow-ups.
The clinic owner now wakes up to confirmed appointments already in the sheet with zero manual effort.
This is what I build for businesses. If you’re still replying manually and losing leads while you sleep — DM me. Let’s fix that. 👇
#MakeAutomation #AIAutomation #NoCode #Makecom #GeminiAI #WhatsAppAutomation #BusinessAutomation #AIReceptionist #FreelanceAutomation #Upwork #Fiverr
Nice build, and very close to a WhatsApp assistant I just built for a dental-clinic use case on the WhatsApp Cloud API. Two things mattered a lot in testing: when a patient asks for a person, the bot has to go completely quiet for that chat until the owner hands it back, and the...
AI Agent vs Body
One AI agent is single-player.
Work is multiplayer.
Start with one agent. I explain it as a body.
The LLM is the brain.
It reads, writes, and reasons. It knows nothing about your business.
RAG is the brain with a library.
It looks things up in your documents before it answers.
MCP is the nervous system.
Model Context Protocol. One standard way to wire the brain to tools and data. Build the connection once, and any agent that speaks MCP can use it.
Skills are muscle memory.
Know-how saved once and loaded when a task calls for it.
The agent is the brain with hands.
It plans the steps, then carries them out in your tools.
My AI assistant runs all day. The expensive model only does the thinking.
Most requests need no AI at all. "Call me in 30 minutes" is handled by plain code. Reminders, schedules and limits are code too.
Short tasks like sorting and scoring go to fast models on a free tier.
The big model only gets planning and long conversations, where it actually has to think.
If it hits a limit, a backup model answers instead of the app going silent.
If your AI app's bill keeps growing, check what really needs the big model.
What does your app send to the most expensive model?
Same pattern holds in outbound tooling. Filtering, dedupe and enrichment lookups are plain code. The model only writes the first line once the row is already verified. Cuts cost and the weird outputs.