Available for AI & Automation Work
I build |
I build AI agents, RAG systems, and automation workflows that ship to production.
The stack I use to design, build, and ship AI systems end to end.
A few things I've built — AI agents, automation pipelines, and machine-learning apps. Tap a card for a closer look.
n8n automation that reads every incoming Packform email, classifies it into one of 11 business events, pulls live order and supplier data from the Packform API, and routes it as a Telegram report, Airtable task, or log entry — running in production.
One dashboard for ad spend, website traffic, and shop revenue for a live e-commerce store, with an AI analyst that answers strictly from the database — built with Next.js, FastAPI, and PostgreSQL, running in production on a VPS.
Always-on n8n agent that reads every incoming support email, classifies it with GPT-5, pulls real order and shipping facts from WooCommerce and QLS, and saves a Dutch reply as a draft for the team to review — running live for an e-commerce store.
A two-engine lead-discovery toolkit that finds people publicly asking for AI automation or web-development help on Facebook and Reddit, scores each one, and serves them through a dashboard and REST API.
Multi-agent systems & orchestration with LangGraph / CrewAI, tool-calling, and human-in-the-loop safety.
24/7 chat and voice receptionists that capture leads, book appointments, and hand off to humans.
Chat over your own documents and databases — accurate answers with citations and guardrails.
Resilient Selenium / Playwright / Puppeteer bots — logins, anti-bot handling, scheduling, and clean structured data.
n8n / Make / Zapier workflows with retries, alerts, and recovery built in — brittle manual ops made self-healing.
From idea to deployed product — Next.js + FastAPI + LLMs, auth, database, the whole stack.
We pin down the real problem, success criteria, and constraints — no scope creep, no guesswork.
I design the agent / automation system and choose the right stack before a line of code is written.
Production-grade implementation with error recovery, logging, and human-in-the-loop safety baked in.
Deployed, documented, and handed over — with monitoring and optional ongoing maintenance.
06About
I'm an AI Engineer focused on turning language models into dependable, real-world systems. I build production LLM platforms, autonomous agents, and full-stack AI tools across CRM, voice automation, and internal-tools domains — shipping features into live platforms used by real teams.
07Contact
Got an AI feature, an agent, or an automation in mind? Tell me the problem — I'll tell you straight whether I'm the right fit.
contactbyabuzar@gmail.com