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
I developed an MVP of a RAG application for querying PDF documents.
The flow is simple: the user uploads a PDF, the system extracts and splits the text into chunks, generates embeddings, stores them in Chroma, and when a question is asked, retrieves the most relevant information so the AI can respond with context.
For this project, I used React, Node.js, Express, LangChain, Gemini, and Chroma Cloud.
The main goal was to better understand concepts such as chunking, embeddings, vector search, and the integration of AI models with vector databases. This is a first version, but it serves as a foundation for more advanced RAG applications in the future.
Amazing work
Thank you so much for commenting!!🙌
You're welcome
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
Hot take: when people say “we can tell you used AI,” they’re usually not reacting to the AI.
They’re reacting to the lack of creative judgment behind it.
I read a post today that explained this sooo well. It basically called out what designers have always been able to tell when someone was using a tool without understanding the craft behind it. That's not new to AI.
You could tell with bad PowerPoints.
You could tell with overdesigned Canva graphics.
You could tell with random stock-photo branding.
AI just lowered the barrier to execution even more, which means taste and direction matter even more now.
Because the difference was never the tool. It was the eye behind it.
Is “taste” becoming the new technical skill?
You could tell with shabby WordPress.
Indeed. You can look at furniture and see that nails and hammer has been used, or not, or you couldn't care less as long as its a good peace of furniture. :)
Recently worked on BES Vinyl - a large-scale ecommerce infrastructure project built around Shopify, FileMaker Server, and AWS to support a rapidly growing online vinyl record business.
The project involved restoring and stabilizing a mission-critical FileMaker PIM system, rebuilding backend workflows, and implementing automated API-driven synchronization between Shopify and the internal product database for inventory, pricing, metadata, supplier imports, and catalog management.
Beyond the storefront experience, the real focus was operational scalability - reducing manual processes, improving data reliability, and creating a centralized commerce ecosystem capable of handling large product volumes efficiently.
Projects like this are where ecommerce, automation, and systems architecture come together to create measurable business impact. 🚀
Everything feels balanced, refined, and professionally designed 💯
😊 AI-Powered Appointment Booking Automation with n8n
This project is an AI-powered appointment booking automation system built with n8n. The workflow is designed to let a user interact with a conversational AI assistant, request an appointment, provide personal details, and move through the booking process automatically.
The system uses an AI agent connected with a knowledge base, memory, vector storage, and external tools. It can understand user messages, ask follow-up questions, collect booking information such as name and email, check calendar availability, create a calendar event, and send confirmation messages through Gmail. The workflow also includes document ingestion, text extraction, embeddings, and Supabase Vector Store integration so the assistant can answer based on stored business information.
My work focused on building the full automation flow inside n8n, connecting AI models, configuring memory and knowledge base retrieval, integrating Google Calendar and Gmail actions, and testing the conversational booking process from start to finish. The result is a practical AI assistant system that can reduce manual appointment scheduling work and provide a smoother customer booking experience.
Skills & Tools:
n8n, AI Automation, Workflow Automation, AI Agent, Google Calendar API, Gmail Automation, Supabase Vector Store, HuggingFace Embeddings, Google Gemini, Groq, RAG Workflow, Knowledge Base Integration, Chatbot Automation, Appointment Booking System, CRM/Business Process Automation.
Appointment booking bot that does the task, nothing makes work easier than this
Trending
Claude
Claude has entered the design space. How are you using Claude Design?
Contra University
Learn from expert creatives how to earn more using next-gen AI tools.
creativeaiflow
Creative AI workflows are evolving. What tools do you use, and what are their strengths and weaknesses?
portfolioreview
The best portfolios tell a story, not just show a grid. Share yours for feedback.
freelancerlife
Freelancer life is wins, pivots, and everything in between. What’s yours right now?