Md Erfan

I’m a backend and AI-focused software engineer with 4+ years of experience building enterprise platforms, AI-powered workflows, and distributed backend systems. I turn complex business requirements into reliable services across .NET and Python, with strong emphasis on RAG/document intelligence and event-driven architectures.

Most recently, I engineered a .NET 10 document intelligence system for HTML documents using smart chunking, sparse vector and dense embedding with Microsoft.Extensions.AI. I also built a private .NET NuGet package for cross-service communication using RabbitMQ event-driven workflows, and led a common background job management system using Quartz for recurring and scheduled event handling.

Previously, I developed a .NET 8 bid management platform with RAG-based AI agent workflows, driving a 40% boost in document analysis efficiency, and built a tax automation system using the GraphBit AI agent framework for auditable, secure sensitive-data analysis. I also delivered a Python multi-agent chatting system using Graph RAG and local LLMs on a clustered GPU setup, and built a CLI pipeline that converts 3TB of document data to text with meaningful metadata—improving vectorisation and model training throughput by 10x.