Browse categories
English
$
USD
Connect to your AI
I will build your rag pipeline or mcp server in python
Adham Momen
About this gig
You need an LLM feature that works on your own data and tools, built so your team can maintain it.
What I build in Python:
- RAG pipelines: ingestion, chunking, retrieval and answers, measured with a small eval set so you know it works
- MCP servers that expose your tools or data to Claude and other agents, with the contract tested
- Backends and automation around LLM APIs (Claude, OpenAI, local Llama models)
What you get: clean, commented source code, tests and a README. Premium adds an eval suite and deployment on your Linux server.
Why me: I built mcp-warrant, which checks whether an MCP server keeps the contract it advertises, and agentarch, a multi-agent RAG system that replays exactly from its trace. I run my own Linux server with Python services, an ops dashboard and unattended automation.
Message me with what the system should do, where your data lives, and how it will be deployed.
Get to know Adham Momen
Adham Momen
AI Red Teaming and LLM Evaluation Engineer
- FromEgypt
- Member sinceJul 2021
Languages
English, Arabic
I test AI systems the way attackers do: prompt injection, jailbreaks, data leakage and unsafe agent tool use. You get a report with working reproductions and fixes. I also build LLM evaluation suites and RAG/MCP backends in Python. I've published tools that red-team agents and audit LLM judges and guardrails (adhammomen.com), evaluate frontier-model code professionally, and contribute to open-source AI projects. I'm a fourth-year medical student and a native Arabic speaker, so I also cover medical and Arabic AI.
My Portfolio
Other AI Development Services I Offer
FAQ
Which models and stacks do you work with?
Python first. Claude, OpenAI and local models; SQLite or the database you already use; Docker on Linux. If you already have a stack, I build inside it.
