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
This is a nicely built n8n automation workflow — an Invoice Parser Agent. Here's a breakdown of what it does:
What it does, end to end:
Watches a Google Drive folder — polls every minute for new PDF files dropped into a specific "Invoice" folder.
Downloads & extracts text — when a new file appears, it downloads it and extracts the raw text from the PDF.
AI-powered data extraction — passes the text to an LLM (via OpenRouter's free tier) configured as an Information Extractor. It pulls out 7 structured fields: Invoice Number, Sender Name, Sender Email, Sender Address, Total Amount, Invoice Date, and Due Date.
Logs to Google Sheets — appends the extracted data as a new row in a Google Sheet called "Invoice."
AI Agent writes a notification email — a second LLM call acts as a "professional email writer," generating a formatted subject line and message body in strict JSON format.
Sends the email via Gmail — fires off the AI-composed email notification to gmpatel404@gmail.com (the billing team).
Error handling — if the AI agent fails, instead of crashing silently, it logs an error message to a separate "Error Sheets" tab in the same spreadsheet for manual review.
Tech stack: n8n + Google Drive + Google Sheets + Gmail + OpenRouter (free LLM tier) + LangChain nodes (Information Extractor, AI Agent, Structured Output Parser)
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
I designed and deployed a custom AI Agent built to handle real operational workloads — not just chat responses. This system combines reasoning, memory, and tool‑use to execute tasks autonomously, reduce manual overhead, and deliver consistent, high‑quality outputs.
What it does:
Processes inputs and makes decisions without supervision
Runs multi‑step workflows with logic and error‑handling
Integrates with APIs, databases, and internal tools
Summarizes, routes, and evaluates information
Operates asynchronously and scales with workload
Why it matters:
This agent replaces repetitive manual tasks, increases operational clarity, and creates a reliable automation layer that founders can trust. It’s built with security, auditability, and real‑world constraints in mind.
Tech used:
LLMs, embeddings, secure tool‑use, workflow orchestration, async operations, and custom logic modules.
Outcome:
A stable, autonomous system that reduces cognitive load, accelerates execution, and frees teams to focus on higher‑value work.
This is the future and most people still aren't paying attention.
Everyone is talking about AI but very few are actually building with it at this level reasoning, memory, tool-use, real workflows. Not a chatbot. Not a gimmick.
A system that actually WORKS while you sleep.
...
A Dubai-based wellness project built around the idea that wellness goes beyond relaxation, it’s about transformation through the senses.
Using lines, circles, and textures, we translated an entire sensory world into a visual identity.
Still one of my favorite worlds I’ve had the chance to build.
Curious, what’s one brand element that made you feel something beyond just liking how it looked?
Brand DesignGraphic DesignGoogle GeminiAdobe PhotoshopAdobe InDesignbranddesignerlogomarkCreative Strategistwellness
this is GORGEOUSSSS woww! 🔥
AI Hiring Workflow & Candidate Automation System
I built an end‑to‑end AI‑powered hiring workflow that automates the entire candidate lifecycle — from screening to onboarding. This system replaces manual review, inconsistent communication, and slow decision‑making with a clean, structured, and fully automated pipeline.
What it does:
• Screens resumes and applications with role‑specific criteria
• Extracts skills, experience, and red flags automatically
• Generates structured candidate summaries for fast review
• Automates candidate communication and follow‑ups
• Schedules interviews and routes candidates to the right stage
• Delivers onboarding steps and training materials automatically
Why it matters:
Hiring is one of the most chaotic operational areas for most teams. This system turns it into a predictable, auditable workflow that saves hours per week and improves candidate experience.
Tech used:
AI models, embeddings, workflow automation (n8n), async operations, secure data handling, custom scoring logic.
Outcome:
A fast, fair, and scalable hiring pipeline that reduces time‑to‑hire, improves clarity, and eliminates repetitive manual work.
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?