I am a Software Engineering graduate from Mekelle University with a passion for designing and building scalable, reliable, and impactful software solutions. I believe that great software engineering starts long before writing code—it begins with understanding real-world problems, analyzing user needs, and designing effective solutions. My approach focuses on domain analysis, defining functional and non-functional requirements, and architecting systems that are modular, maintainable, and scalable. I enjoy transforming ideas into high-quality applications that solve meaningful problems and create value for users while continuously learning and growing as a software engineer.

• AI / Machine Learning (applied) — Production RAG, classification, and predictive performance analysis.
• Backend Architecture — Scalable APIs, Modular Monolithic systems, and High-Concurrency design.
• AI-driven Products — Healthcare, Education, FinTech (Payments), and Automation.
• Productization & Integrations — Payment flows (Chapa), real-time notifications (SMTP/Redis), and Cloud deployments.

Integrations: Payment flows (Chapa integration), Celery Background Tasks, SMTP Mail, Geocoding & Mapping.

The Problem: Modern e-commerce often suffers from "monolithic bloat," where updating a single feature risks breaking the entire system. Businesses lose time, money, and energy dealing with slow deployments and unoptimized background processing.

The Initiation: This project was initiated to build a future-proof retail engine. By architecting a modular monolithic system with strict domain separation, I ensured the platform could scale effortlessly. I integrated Redis/Celery for high-speed background tasks and Locust for stress testing to ensure the system doesn't just work, but thrives under pressure.

• Backend: Django REST Framework (DRF) + Redis + PostgreSQL + SMTP.
• Frontend: Next.js + Tailwind CSS (Deployed on Vercel).
• Key Achievements: Modular domain design (Auth, Orders, Payments), Pytest coverage, and Locust performance benchmarking.
• 🔗 Live Demo | Repository

The Problem: Many schools still rely on manual, error-prone administrative workflows. This "traditional gap" leads to inaccurate attendance, delayed financial reporting, and a lack of visibility for parents regarding their child's academic trajectory.

The Initiation: The goal was Digital Transformation. I designed this system to bridge the gap between admins, parents, and students. By implementing QR-based attendance and an AI performance-prediction engine, I transformed a passive record-keeping tool into an active, predictive educational assistant.

• Stack: React + TypeScript + Vite (Frontend) | DRF (Backend).
• Architecture: Transitioning to Go (Gin) for high-concurrency payment flows and FastAPI for AI performance.
• Innovation: AI-driven early performance analysis and automated financial invoicing via Chapa.
• 🔗 [Repository Link] (Coming Soon)

The Problem: Early education is often one-size-fits-all, ignoring that every child learns at a different pace. Without personalized feedback loops, children can easily fall behind or lose interest in tutor materials.

The Initiation: This project focuses on Adaptive Learning. I initiated a role-based ecosystem where the system itself acts as a digital tutor. By analyzing kid-specific performance data, the backend dynamically recommends videos and tests, ensuring the educational content is always perfectly matched to the child’s current level.

• Stack: Django REST Framework (API) | Flutter (Mobile App).
• Core Feature: Fine-grained RBAC for Teachers, Parents, and Kids + AI Recommendation Engine.
• 🔗 Repository

The Problem: Parents frequently struggle to accurately track their child’s nutritional health and identify the caloric/nutritional value of daily meals. General advice is often too broad to be useful for specific weight/height milestones.

The Initiation: I combined Computer Vision with RAG (Retrieval-Augmented Generation) to solve the information gap. This platform allows parents to get instant nutritional breakdowns just by snapping a photo of a meal. By syncing this data with the child's physical growth metrics, the AI provides a tailored "nutrition roadmap."

• Stack: Python, FastAPI, RAG Frameworks, Computer Vision.
• Key Innovation: Image recognition for meal content analysis and educational RAG-based chatbots for instant parent support.
• 🔗 [Repository Link] (Coming Soon)

• Internship-Portfolio: Verified professional work at Medco Technology.
• Framework-Evaluations: Technical benchmarks comparing DRF vs. Gin vs. FastAPI.
• University-Archive: 4 years of academic engineering evolution.