Skip to content
View aelmasry's full-sized avatar
🏠
Working from home
🏠
Working from home

Block or report aelmasry

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
aelmasry/README.md

Ali Salem

Software Solutions Architect | Applied AI & Automation

I’m a Solutions Architect and Senior Software Engineer with 18+ years of experience building backend systems and distributed applications used in production. Over the years, I’ve worked across telecom, SaaS, and retail, designing systems that are reliable, scalable, and built to last.

I’m currently at Seddiqi Holding in Dubai, where I work on enterprise retail systems used across the GCC. I own four projects covering a mix of backend development, legacy POS modernization, mobile integrations, and production operations. Alongside my day-to-day work, I build internal AI and automation tools, including MCP servers, and spend time exploring RAG and LLM-based applications through side projects.

I work primarily in Python and TypeScript, and at a high level with Go and Java. I also maintain and modernize large PHP platforms, which is where much of my earlier production work sits. I enjoy designing system architectures, building APIs, integrating different platforms, and improving how software is delivered through CI/CD, observability, and automation. I like solving problems that make systems easier to operate and easier for other engineers to maintain.

Before Seddiqi: Technical Lead at TechieMatter, led two teams at Serv5 (bugs down ~40%), built Lamha's APIs, and worked on Resalty.net at 1M+ SMS/day across MENA.

Selected work
  • local-rag-ollama — self-hosted RAG pipeline: ingest, chunk, embed, retrieve, and answer locally with Ollama, FAISS, and Docker
  • NestAgent — Python agent harness built without LangChain or LangGraph, to understand orchestration from first principles
  • TaskMind — multi-provider AI agent application in TypeScript

I write up the architecture decisions behind this work as decision records: when a local LLM is required instead of an API, when GraphRAG earns its complexity over plain RAG, and when Kubernetes is over-engineering. For me, the interesting part of AI is the judgment calls, not the frameworks.

If you're interested in software architecture, backend engineering, engineering leadership, or practical AI, I'd be happy to connect.

Languages

Python TypeScript JavaScript Go Java PHP Bash SQL

AI & Automation

Ollama FastAPI LangChain Hugging Face Streamlit RAG MCP

Frameworks & Platforms

NodeJS ExpressJS NextJS GraphQL Laravel Symfony VueJS AngularJS

Cloud & DevOps

Docker GitHub Actions AWS Azure NGINX Apache Linux

Database

MySQL PostgreSQL MongoDB Redis Elasticsearch

Others

Git GitLab WordPress Magento

Currently exploring

Kubernetes Terraform LangGraph Ragas

Pinned Loading

  1. local-rag-ollama local-rag-ollama Public

    Python

  2. NestAgent NestAgent Public

    Python

  3. mystack mystack Public

    MyStack - Complete VPS Docker Environment

    Shell

  4. TaskMind TaskMind Public

    TaskMind is a local AI Agent application designed to transform any user prompt into a structured task. Instead of simply answering questions, TaskMind treats each prompt as a goal, breaks it down i…

    TypeScript 1

  5. Ollama-AI-Assistant Ollama-AI-Assistant Public

    VS Code extension that integrates with locally hosted Ollama models for code improvements and assistance create by Ali Salem

    TypeScript