We’re building AI-powered back-office automation for convenience-store and gas-station operators, starting with invoice processing, vendor reconciliation, pricebook matching, exception review, and workflow automation.
This is a hands-on founding role for someone who likes building real production systems around messy documents, APIs, workflows, databases, and AI agents.
You may be a fit if you have experience with:
Backend or full-stack engineering
Python, TypeScript/Node, APIs, databases, queues, and cloud infrastructure
Workflow automation tools such as n8n, Zapier, Make, or custom automation systems
LLMs, document AI, OCR, RAG, evals, or agentic workflows
Shipping practical systems for real business users
Startup-level ownership and willingness to work closely with founders
Location: Houston/Texas preferred. Remote possible for an unusually strong fit. Compensation: $104k–$165k cash + bonus + meaningful equity. Role type: Full-time founding role preferred. Open to starting with a paid project/trial if that makes sense.
Apart from that I’m also a full-stack developer with the right Gen AI experience, which makes me a solid plus for your team[but right now only vibecoding]
Check my recent gen ai projects… I built a native Android automation agent too. It’s worth a look:
I can build complex AI automations directly in code, not just inside n8n
I recently started posting my n8n work on YouTube with explanations:
Hi, I’m an automation systems builder based in the Philippines. I specialize in building end-to-end production workflows using n8n, APIs, webhooks, and AI integrations.
I’m not a traditional backend engineer, but I build practical systems that run businesses without breaking quietly. Got anything for me?
I’ve spent the last 2 years building production-grade backend systems and AI agents that thrive on unstructured, messy data. When I saw you were tackling back-office automation for convenience stores and gas stations, it immediately clicked. DSD (Direct Store Delivery) invoices, complex price book matching, and fragmented vendor portals are a massive, unglamorous data problem that is absolutely ripe for a combination of Document AI and agentic workflows.
I’m reaching out because I want a hands-on, high-ownership role where I can bridge the gap between cutting-edge LLMs and real-world business value.
Core Competencies & Technical Alignment
Here is how my experience maps directly to your technical stack and roadmap:
AI, OCR & Document Processing: I have hands-on experience extracting structured data from heavily skewed, faded, or handwritten documents using [mention specific tools, e.g., Google Document AI, AWS Textract, or GPT-4o/Claude 3.5 Sonnet vision capabilities]. I understand how to build robust evaluation pipelines (evals) to prevent hallucinations in financial data.
Workflow Orchestration & Automation: Deeply familiar with building event-driven pipelines. I have built complex, multi-step agentic workflows using [n8n / Zapier / Make], integrating human-in-the-loop (HITL) exception handling for when the AI confidence score drops below a specific threshold.
Backend & Cloud Infrastructure: Strong foundation in [Python / TypeScript / Node.js]. I design resilient, scalable APIs and manage asynchronous queues (e.g., Redis, Celery, AWS SQS) to handle heavy document-processing workloads without timing out.
RAG & Database Architecture: Experienced in vector databases (e.g., Pinecone, Weaviate) for semantic search and traditional relational databases (PostgreSQL) for maintaining single-source-of-truth pricebooks and vendor ledgers.
What I’ve Built (Portfolio & Links)
1. Automated Invoice Reconciliation Engine]
What it does: Built an end-to-end pipeline that ingests messy PDF invoices via email, extracts line items using OCR + LLMs, and flags price discrepancies against a master database.
What it does: Developed a workflow that fuzzy-matches messy vendor SKU descriptions with clean internal pricebook records using vector embeddings, reducing manual reconciliation time by [X]%.
Location: [Your City/State] (If not Houston: I am fully set up for high-communication remote work and am willing to travel to Houston periodically for intensive sprint planning and alignment.)
Availability: I am available to start immediately
I saw you are open to starting with a paid project or trial. I am highly receptive to this I believe the best way to prove we are a strong fit is to get our hands dirty in a shared codebase and ship a small feature or workflow together.
Looking forward to the possibility of building Octane AI together.
I am Muhammad Bin Zohaib, an AI & Web Solutions Specialist, Full‑Stack Developer, and Certified n8n Developer (Level 1 & 2). I specialize in building production‑grade automations, voice agents, and intelligent backend systems with error handling, webhook validation, and complex multi‑branch logic.
I have delivered automation systems for clients across Canada, Greece, Singapore, India, Sudan, Spain, Australia, Germany, and the UK. Relevant experience includes:
AI Appointment Setter (Google Calendar + WhatsApp + OpenAI) – automated lead qualification and booking.
I am open to aligning on a monthly fee plus implementation fees, and I am confident in delivering workflows with idempotency, HMAC validation, and error handling as per your spec. I can also provide Loom walkthroughs for each workflow group.
This matches what I have been building for the past three years.
At Velocity BPA, I built a Zero Human Company platform: n8n workflows and Claude AI handle opportunity discovery, code generation, publishing, and client management end-to-end. The pipeline generates complete npm packages autonomously, 162 published so far, using parallel Claude API calls, GitHub Actions CI/CD, and Airtable for project tracking. I also built the Contractor AI Agent Platform, a mobile-first application in which Claude handles the full contractor customer lifecycle: intake, bidding via RAG on previously accepted bids, scheduling, invoicing, and follow-up. Stack is AWS ECS Fargate, Aurora PostgreSQL, DynamoDB, Qdrant, AssemblyAI, and n8n for AI orchestration.
The invoice processing, vendor reconciliation, and pricebook matching problems you are describing, messy documents, APIs, and agentic workflows working together, are exactly the kind of systems I build. I work in TypeScript/Node.js primarily, with Python where it fits, and have been shipping production AI systems since early 2024.
Core Competencies & Technical Alignment
What I’ve Built (Portfolio & Links)
Quick Links:
Logistics & Availability