Software Engineer · Full-stack delivery

Senior software engineer. I turn ideas into deployed products, fast.

I own delivery end to end — Python & FastAPI/Django on the backend, TypeScript & React on the front — and get production systems live in the minimum time. Speed from how I work, not from cutting corners.

See selected work Start a conversation

Years shipping

10+

Platforms shipped

90 %

Faster load times

40 %

Efficiency via async

About

Four years turning ideas into systems that stay up.

My edge is delivery. I take an idea from a first conversation to something running in production in the minimum time — designing the backend, building the React frontend, and owning the path between them. Across 10+ platforms, that ownership is what's kept timelines short.

Fast doesn't mean fragile. The speed comes from the way I work: clear boundaries between layers, async and scheduled tasks doing the heavy lifting off the request path, and decisions made once and reused — not from skipping the parts that keep a system reliable.

What "production-grade" means to me

Observable

You can see what it's doing in production — and know before users do when something's wrong.

Secure by default

Auth, access, and data handling are decided up front — JWT, scoped permissions, no afterthoughts.

Maintainable

Structured so the next change is cheap — Feature-Sliced frontends, clear backend boundaries.

Efficient at rest

Background and scheduled work keeps the request path fast and the system lean over time.

Selected work

problem → approach → result

Flagship · Enterprise SaaS

Agentic AI Blueprint

An internal Deloitte platform, built end to end — from idea to production agents running across the enterprise.

PythonFastAPITypeScriptReactSSE

Problem

Enterprises needed a structured, fast way to plan AI transformation — roadmaps, workforce-impact analysis, and deployable agents — grounded in real industry processes rather than guesswork.

Approach

Owned the full stack. A real-time LLM streaming layer over a Server-Sent Events pipeline drives multi-phase taxonomy generation, so results render as they're produced. A strict Feature-Sliced Design frontend keeps the codebase scalable and maintainable as scope grows.

Result

Generates AI transformation roadmaps, workforce-impact analysis, and production-ready agent deployments — backed by a curated taxonomy of industry processes.

Enterprise platforms

1,100+

Industry processes

Industries covered

E-commerce

KMC

A storefront for dental tools, built to stay fast as the catalog grew.

DjangoPostgreSQLCelery

Problem

A growing catalog and media-heavy pages were dragging load times and hurting the buying experience.

Approach

Restructured data access, offloaded heavy work to Celery background tasks, and tuned caching across the stack.

90%

Faster load times across the storefront.

Fitness platform

LiveIt8

An app connecting trainees with certified trainers, kept lean by design.

DjangoDRFJWT

Problem

Unused and stale data was accumulating, adding operational overhead for trainers and trainees.

Approach

Built a secure REST API with JWT auth and scheduled Celery jobs that automatically prune unused data and keep the system lean.

40%

Operational-efficiency gain via async & scheduled tasks.

Real estate

Iconic

A company profile that editors can keep fresh — and that search engines can read.

DjangoDRFSEONginx

Problem

A static site couldn't keep listings and content fresh, or rank well in search.

Approach

Built a fully dynamic, SEO-aware CMS with Django & DRF, served behind Nginx for performance.

A self-served, search-friendly CMS — editors publish SEO-optimized content dynamically.

Contact

Have something to ship? Let's talk.

Freelance project or a role on your team — tell me what you're building and the timeline you're working against.

→ contact@youssefshaker.com → linkedin.com/in/yousef-shaker → github.com/joeashaker