Alireza Mohagheghi
Senior AI & Full-Stack Engineer · Production AI Agents
- 3
- Followers
My voice agent's median response time went from 1.7 s to 1.08 s without changing the model, with every part of the stack running in Europe.
I'm building an AI receptionist that answers phone calls for businesses in Cyprus in Greek, English and Russian, and books appointments into...
Prototypes are easy now. Production isn't.
I just relaunched my site, alireza.work, a one-page look at what I build and how I work.
I'm a senior full-stack & AI engineer in Cyprus. Over the last 7 years I've built:
→ a multi-tenant commerce platform running ~300 stores, with an AI...
I got tired of looking at used cars and trying to figure out whether a listing was actually a good deal.
So I built Carcy.
You give it a car listing, and it analyzes the price, specs, history and potential red flags, then gives you a report.
I built it 5 days ago. 46 reports later,...
A year and a half of architecture decisions, plugin systems, and shipping an agentic commerce platform with a two-person engineering team.
We landed on a kernel-based, modular design. The kernel holds only data structures, entity management, and tenant management, nothing else....
medium.com
Zero to Platform: What Building as Full Stack Engineer Actually Looked Like
A year and a half of architecture decisions, plugin systems, and shipping an agentic commerce platform with a two-person engineering team.
How I reduced our AI agent's input context by ~90%.
Before optimizing MCP in Webito, our agent received all 96 tools and their schemas upfront — around 13–14K input tokens before it could even start working.
I changed the MCP architecture so the agent first discovers available...
