IMRaD Introduction Classification SaaS
In my engineering thesis. 8-node microservices architecture using Gemini Pro + BERT for rhetorical classification of scientific papers. 169k-sentence dataset, 98.21% peak F1.
Software Engineer
Software engineer with 5 years in full stack web and mobile development, a proven track record in data science, research, and modern AI solutions, with a business-first engineering mindset for shipping products that deliver real value.
5 years building production systems across the full stack: REST APIs, relational databases, responsive frontends, and cross-platform mobile apps shipped and maintained at scale.
Research-grade ML and NLP background. Engineered a 169k-sentence dataset and deployed an 8-node AI microservices platform achieving 98%+ classification accuracy in production.
Business-first mindset with end-to-end ownership. I scope requirements, drive architecture decisions, and ship every feature autonomously, treating every product as my own.
Product Engineer
bareedbox LLC · United Arab Emirates (Remote)
Sole technical owner across the full product lifecycle: scoped requirements, drove architecture, and shipped every feature end-to-end autonomously.
Architected a 110+ model multi-tenant logistics hub, providing direct order synchronization for merchants and dedicated fulfillment tools for shipping companies.
Managed a 4-person development team across 2 production codebases, handling technical interviews and standardizing code reviews.
Built an eCommerce sync engine for Shopify and WooCommerce, implementing the Strategy pattern to allow pluggable, non-breaking store connections.
Integrated Stripe Connect to build financial infrastructure, supporting merchant and carrier wallets, automated payouts, and on-demand withdrawals.
Backend Engineer
Addinfo · Paris, France (Remote)
Created a high-throughput component statistics scan ingestion engine that converts huge and complex JSON snapshots into 50+ PostgreSQL relational tables.
Decreased analytics latency from 3s to <100ms by using Write-time Aggregation (Materialized View approach),application-level caching, and SQL query optimization.
Monitored, debugged, and fixed memory leaks across 100+ live production endpoints by using flame graphs and implementing a global heap interceptor.
Created a multi-tenant backend with multi-provider OAuth (Google, GitHub, GitLab, Azure).
Backend Software Engineer
CreaTech dz · Algiers, Algeria
Developed multi-tenant Events Marketplace backend with Twilio SMS-OTP onboarding, RBAC guards, partner approval workflow, and geo-tagged discovery for organizers and sellers
Built out full revenue engine: Stripe webhook payments, frequency based loyalty system, SaaS tier subscriptions, moderator permissions, CMS (blogs and guides), and docker CI/CD and automated cleanup
Developed Ecole at Home LMS backend automating virtual-school operations: tri-role identity, teacher availability mapping, and RxJS-resilient Zoom scheduler removing manual bookings
Built out the entire student experience from start to finish: Stripe payments, automated Zoom-to-Vimeo recording pipeline, transactional emails, and dedicated upcoming and past course portals
Full Stack Engineer Intern
Data Intuition · Algiers, Algeria
In my engineering thesis. 8-node microservices architecture using Gemini Pro + BERT for rhetorical classification of scientific papers. 169k-sentence dataset, 98.21% peak F1.
A cross-platform vocabulary learning app deployed on web and android, built with TypeScript, Expo and Domain Driven Design. Features include spaced repetition reviews (SM-2), hands-free podcast narration, bilingual (EN/FR) word tracking with definitions and context sentences, and offline-first backup with ZIP-based sync — all from a single codebase with zero cloud dependency.
Avoid paying a fortune to the cloud just to display your portfolio. Learn how to mock API calls in Next.js to deploy complex, microservices-based resume projects for free.
In this reflection, I share how I completed both my master's thesis and engineering project on IMRAD classification. From building a 169k-sentence dataset and fine-tuning BERT models to a 98.21 percent F1 score, to launching a full microservices SaaS platform. This post details the complete technical journey.
Master Python asyncio: learn coroutines, async/await, the event loop, Tasks, asyncio.gather, TaskGroups (Python 3.11+), and Futures for non-blocking I/O.
I'm currently open to new roles, remote-first or international. If you're building something ambitious, I'd love to hear about it.