Backend to AI agents
I'm Reza Shirazi, Head of Engineering at a US B2B SaaS company, where I lead a small team. I build backend services, websites, search, cloud and AI agents that run every day.
You bring the problem. I plan every layer it needs.
I build each layer, from cloud to AI agents.
All six work as one. I keep them running as you grow.
Picture: a neat block of small boxes rises and moves into six layers, stacked one on top of the other. From bottom to top: DevOps and cloud, data and search, backend services, admin console, website platform, and MCP and AI agents. Links join the boxes in each layer and the layers next to each other, and small lights move along the links. One amber box sits on every layer, in one column: amber shows where I work, on every layer.
Production platforms, from the database to AI agents.
Agents that do real admin work, with a confirm step before any change.
Read the case →One code base that runs many customer websites and their admin tools.
Read the case →The daily web app for staff. Each person can do only what their role allows.
Read the case →Python services on one shared framework I created.
Read the case →Data comes in from many feeds, gets cleaned, and is searchable fast.
Read the case →The whole platform written as code, moved to Kubernetes with a tested plan.
Read the case →Work at a US B2B SaaS company, by area.
A small team runs many services the same safe way, with people and AI side by side.
I designed and built, alone, the AI that staff, customers and shoppers use on live data.
Read the case: AI agents on live business data →Many customer websites run from one code base, and staff edit pages with no developer.
Read the case: Multi-tenant website platform →The platform and all its data moved to a new cloud in three planned nights, nothing lost.
Read the case: Platform move to Kubernetes →Right offers, rates and monthly payment plans on every customer site, set by staff rules.
Read the case: Pricing, offers and rule engines →I wrote the first backend in 2023, then led its move into separate services. The apps saw no change.
Read the case: Platform backend and shared framework →One web app where operations staff do daily work: accounts, billing, leads and support.
Read the case: Staff console →03 Path
From freelance work to leading engineering. Each step added new skills.
Websites and backend work for clients.
Python backend services, including a public search API.
I lead a small team and still write much of the code.
Hiring an engineering lead, or building a platform or AI agents? Tell me about it.