Built a fully automated Expense Request Approval system using Airtable, Fillout Forms, and Make.
Key outcomes:
• Captures employee expense submissions and creates Airtable records automatically.
• Notifies managers via email for approvals and updates request status dynamically.
• Alerts finance officers and logs all approvals or rejections.
• Eliminates manual tracking, streamlining approvals with event-driven automation.
• Provides transparency and reduces errors across the finance process.
0
19
Built a production-ready RAG AI Agent for a real estate company using n8n, Supabase (pgvector), and Open AI.
The system ingests documents from Google Drive, chunks and embeds them into a vector database, and powers a Telegram chatbot named "Collin" that:
• Answers client queries in real time
• Retrieves property listings and sends images
• Collect user details (name, email, and phone number) as well as properties requested or discussed during the conversation.
• Automatically save all collected data in Google Sheets as LEADS
• Provides a smooth, automated client experience
0
17
Built a custom Airtable frontend interface and dashboard that visualizes key inventory metrics in real time.
Key outcomes:
• Track stock levels and monitor inventory trends
• Analyze product performance at a glance
• Provide live, actionable insights for faster operational decisions
• Eliminate manual spreadsheet tracking and reduce errors
0
19
Built an end-to-end automated loan processing system for a private money lending firm using n8n, Airtable, and GPT-4o. Designed a custom Airtable interface for managing loan applications, eligibility criteria, and approved records. An AI Agent evaluates each applicant against a dynamic criteria table, calculates interest and total repayment amounts, and triggers branded HTML email notifications for approvals and declines. Approved loans follow a confirmation workflow with reminder logic, all running without human intervention from application to final decision.