I take products from nothing to real users in production — and then own them there. Five years of backend and full-stack engineering: mission-critical industrial platforms, fintech systems in regulated environments, and two platforms shipped from zero to daily active use.
Python-first with FastAPI and Django, strong on React and TypeScript, and hands-on with the infrastructure underneath — Docker, Kubernetes, Terraform. Most recently building production AI: LLM integration, fine-tuning and agentic tooling.
Ahmed Eltawil
SENIOR SOFTWARE ENGINEER
Engineering experience
Systems shipped
Platforms taken to production
Certifications
6 production systems across SaaS platforms, real-time industrial infrastructure, live-capital fintech and AI tooling. Each one covers the constraints it had to survive and the architecture chosen to meet them.
ProductionFull-Stack2024 — Present
Multi-tenant SaaS platform taken from idea to live product, built solo
ProductionReal-Time2023 — 2026
Full software stack for a mission-critical ADNOC robot, architecture to field operation
2022 — 2023
Live-capital trading backend with broker integrations and user-configurable bots
2023
Full student lifecycle platform delivered freelance, client meeting to production
2020 — 2022
Enterprise system across five departments, plus a sensor-to-dashboard monitoring pipeline
More on the way
Demos and open-source work are being written up. Reach out if you'd like detail on something specific in the meantime.
I started in Mechatronics and Automation Engineering, and found the interesting problems were all on the software side.
Not the mechanism — the system around it. How state stays consistent when four departments write to the same record. How a control channel behaves when the network drops mid-command. How you keep a query fast once the table has a few million rows in it. Those turned out to be the questions I wanted to spend my time on, so I moved into software properly and haven't looked back.
Five years on, that's been backend architecture in Django and FastAPI, real-time WebSocket infrastructure, and the databases and Kubernetes pipelines that hold it all up. The domains have varied — a manufacturing ERP, a live-capital trading platform, a KYC product in a regulated environment, an autonomous industrial system at ADNOC — but the work is consistent: design the system, make it correct, then make it survive production.
What I keep coming back to is the zero-to-one part. Twice now I've taken a product from a blank page to real users paying attention — most recently SayTech, a multi-tenant SaaS platform I built solo, down to the Terraform and a fine-tuned model serving natural-language queries over live business data.
The engineering background still earns its keep. When the backend talks to physical hardware, knowing what happens on the other side of the wire is the difference between a clean integration and a long week.
Education
B.Eng Mechatronics & Automation
Ain Shams University — graduation project A+
Exchange
Electrical Engineering
California State University, Los Angeles
Based in
Abu Dhabi, UAE
Open to remote and on-site
Languages
Arabic, English
Professional working proficiency
Scalable API design in FastAPI and Django, distributed microservices with Redis pub/sub, JWT/RBAC security, and the background job pipelines that keep request threads free.
Taking a product from requirements through architecture, build and deployment to real users — twice single-handedly. Including the unglamorous parts: handover docs, user training, client integration.
LLM integration against Anthropic and OpenAI APIs, fine-tuned tool-calling models served as their own microservices, agentic tooling, and MCP servers for assistant-accessible data.
WebSocket infrastructure for bidirectional streaming — concurrent sensor telemetry at sub-second latency, live operator dashboards, and remote control channels that fail safely.
Schema design and query optimisation across PostgreSQL, MongoDB and MySQL — including indexing work that cut API response times ~40% — with Redis for caching and Elasticsearch for search.
Kubernetes on Terraform-managed cloud, CI/CD pipelines that took deployments from days to under 30 minutes, health checks, structured logging and Prometheus/Grafana monitoring.
Five years of backend and full-stack engineering across industrial autonomy, fintech, regulated compliance and manufacturing.
Owned the full software stack for an autonomous industrial robot on a mission-critical ADNOC programme — from architecture through deployment to live field operation on an ADNOC site.
Grouped by what it does rather than by label. The 39 highlighted items are the ones I use in production regularly — the rest I'm competent in and reach for when the problem calls for it.
Service design, API surfaces and the work that happens off the request thread.
LLMs in production — integrated, fine-tuned and given tools, not just prompted.
I build and own the interfaces that sit on top of my APIs.
Schema design, query performance and caching strategy.
Getting it deployed, keeping it up, and shipping the next change safely.
Where software meets physical equipment — the background that makes those integrations quick.
coreused in production regularly
15 completed certifications weighted toward infrastructure, backend and architecture, plus 2 currently in progress. Completed credentials link out for verification.
DevOps
KodeKloud · In progress
DevOps
Manara · In progress
DevOps
Udemy · Oct 2025
DevOps
LinkedIn Learning · Completed
DevOps
PagerDuty & LinkedIn · Completed
Backend
LinkedIn Learning · Completed
Backend
LinkedIn Learning · Completed
Architecture
LinkedIn Learning · Nov 2024
Architecture
LinkedIn Learning · Nov 2024
Frontend
LinkedIn Learning · Oct 2022
AI/ML
LinkedIn Learning · Feb 2022
AI/ML
LinkedIn Learning · Jan 2022
AI/ML
LinkedIn Learning · Dec 2021
Management
LinkedIn Learning · Completed
Management
Coursera (Google) · Completed
Management
Dassault Systèmes · Completed
Available immediately for senior engineering roles and consulting — backend, full-stack, or AI engineering. If you have a product to take from zero to production, or a hard problem already there, I'd like to hear about it.