Dawood Saleh is a software engineer based in Sana'a, Yemen, specializing in backend systems, serverless architecture, and AI tooling. He has worked independently with international clients since 2023, building backends on Cloudflare Workers, Telegram automation bots, and tools that connect AI models to real-world data through the Model Context Protocol (MCP).

MCP is an open standard that lets AI models call external tools and query live data sources directly, instead of relying only on what they were trained on. It's a core focus of this work because it solves a real problem: giving AI assistants reliable, structured access to specialized data. The MCP server built for Arabic academic research is a concrete example — it gives AI models direct, structured access to over 4 million previously unindexed Arabic research records from Mandumah, ASJP, and SHAMAA.

The primary stack is Node.js and TypeScript, deployed on Cloudflare Workers for serverless backends. Beyond MCP servers, this includes Telegram bot development (including a media downloader supporting 9+ platforms), Cloudflare KV for storage, webhook-based automation, and scheduled cron tasks — all built to run with minimal ongoing maintenance.

Arabic (native) and English (professional working proficiency), which makes it possible to work directly with both Arabic-speaking and international clients without translation friction.

Beyond independent engineering work since 2023, Dawood has taught Object-Oriented Programming at Hajjah University and written technical research for academic and engineering audiences. That teaching background shows up in how projects are documented and explained — clearly, and without unnecessary jargon.

Both. Dawood is open to freelance/contract engagements as well as full-time backend engineering roles, particularly where the work involves AI tooling, serverless architecture, or MCP integrations.

The fastest way is email: eng.dawoodsaleh@gmail.com. Include a short description of the project or role, and expect a reply with follow-up questions or next steps.

Yes. The blog publishes a standard RSS feed at /rss.xml, so you can follow new posts about backend engineering, AI tooling, and MCP in your feed reader of choice without checking the site manually.

All blog content on this site is all rights reserved. Short quotes with clear attribution and a link back to the original post are generally fine, but full republishing or redistribution requires explicit permission first — reach out via email to ask.

Projects involving backend systems on serverless infrastructure (especially Cloudflare Workers), AI tooling that needs to connect language models to real data via MCP, or automation built around messaging platforms like Telegram. If a project needs reliable infrastructure that runs without much hand-holding, that's the sweet spot.