Skip to content
View Eslam93's full-sized avatar
🏠
Working from home
🏠
Working from home
1 follower · 0 following

Achievements

Achievement: QuickdrawAchievement: Pair ExtraordinaireAchievement: Pull Sharkx3Achievement: YOLO

Achievements

Achievement: QuickdrawAchievement: Pair ExtraordinaireAchievement: Pull Sharkx3Achievement: YOLO

Block or report Eslam93

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
Eslam93/README.md

Eslam Hamed - Technical Lead and AI Solutions Architect. Production agentic AI, RAG systems and cloud-native platforms on AWS and Azure.

Open to new roles - let's talk eslamhamed.com LinkedIn eslam1311@outlook.com

About

I've spent 10+ years in software engineering, working across Python, TypeScript and C#/.NET, and I ship cloud-native AI platforms on AWS and Azure.

I've shipped production AI to enterprise customers, including Gulf-region government ministries. I formally led a 5-person cross-functional team through a startup's growth phase, and I ship full-stack products solo. Right now I'm focused on production agentic AI, RAG at scale, LLMOps, and AI-assisted engineering done with real discipline.

Selected work

Verifiable AI analyst case study

Verifiable AI analyst

Every number carries a receipt. Executives ask in plain language. The model decides what to ask, a deterministic engine does the math, and any claim without a receipt is rejected. Solo build, 2026.

 Auto-RAG and MerlinX case study

Auto-RAG and MerlinX

At DXWand: enterprise retrieval in production for 7-10 customers, then the agentic SaaS on top of it. Cut LLM cost per conversation by 73% across 840 agents, proven on a frozen suite at 95% confidence before rollout.

 Career Rafeeq case study

Career Rafeeq

A paid AI SaaS I built and run solo. A statistical eval harness gates every prompt change, and 53 integration test files gate every deploy. Live at careerrafeeq.com.

Open source: agentic-methodology - the Claude Code harness I built the analyst with. Rules, skills, hooks and evidence-dated facts, derived from real installations.

Earlier work worth knowing about

  • Egyptian Ministry of Justice - real-time speech-to-text with speaker identification for the court system, selected by the Minister and deployed across several dozen courtrooms.
  • Egyptian Cabinet IDSC - semantic legislation search, delivered as prime vendor from pre-sale through handoff.
  • Cogent - an Arabic-dialect NLP chatbot SaaS that handled 100K+ monthly sessions at 91% intent accuracy (since retired).

What I do

  • AI and LLM systems - production RAG, multi-agent systems with tool calling, orchestration with LangGraph, LangChain and LlamaIndex, MCP tool gateways.
  • AI operations and safety - LLMOps, evaluation harnesses, guardrails, verifiable-evidence architectures, human-in-the-loop review.
  • Cloud-native backends - services in Python, TypeScript and C#/.NET; microservices, event-driven and serverless on AWS and Azure; release engineering and repository governance at estate scale.
  • Full-stack, solo - strict TypeScript with React or Svelte, owning a product end to end.
  • Engineering leadership - team management, hiring, mentoring, architecture and code review, engineering standards.

Stack

Code Python, TypeScript, C#, .NET, FastAPI, Node.js, React, Svelte, GraphQL
Cloud AWS, Azure, Docker, Kubernetes, GitHub Actions, Kafka, RabbitMQ
AWS: EKS, Bedrock, CDK, Lambda  |  Azure: AKS, Functions, Service Bus, Azure DevOps
Data PostgreSQL, Redis, MongoDB, Elasticsearch
Also SQL Server and Cosmos DB
AI LangChain, LangGraph, Model Context Protocol, Neo4j, Temporal, n8n
LangChain, LangGraph, MCP, Neo4j, Temporal and n8n, plus LlamaIndex, Weaviate and Azure AI Search
Builds with Claude Code, GitHub Copilot
Claude Code, Codex and GitHub Copilot, every day

Based in Cairo, working remotely  |  Arabic (native), English (full professional)  |  B.Sc. Computer Science, Helwan University

Pinned Loading

  1. Murshid Murshid Public

    Arabic-first RAG over Saudi government-services FAQs — 5-provider bench (Mock/Claude/OpenAI/Gemini live + Falcon-Arabic stub), structured-output judge, red-team adversarial harness, deterministic p…

    Python

  2. A.I.R.A A.I.R.A Public

    Autonomous investment-research backend in .NET 8 — async job pipeline with observable progress, structured reports, and report-grounded chat over SignalR via the OpenAI Responses API.

    C#

  3. agentic-methodology agentic-methodology Public

    A Claude Code harness and a knowledge-base format for building software with an AI assistant: rules, skills, hooks, and evidence-dated facts, derived from real installations.

    Shell

  4. Eslam93 Eslam93 Public

    Profile README

  5. mini-library-management-system mini-library-management-system Public

    A library catalog and circulation system with an AI Copilot that works on top of the product's real capabilities

    Python

  6. SignalDesk SignalDesk Public

    Channel-to-ticket triage workspace with human-in-the-loop review (FastAPI + Vite/React + LLM tool-calling)

    Python