Building AI that understands financial risk
Computer Science student and backend developer building secure, scalable APIs and ML pipelines for Ethiopian and international fintech.
Summary
Computer Science student and backend developer with a strong focus on machine learning for financial systems. Skilled in building secure, scalable backend APIs with Django, and in designing ML pipelines that turn raw transaction, review, and complaint data into credit risk, fraud, and NLP models with measurable accuracy.
Comfortable working across the full ML lifecycle — from data cleaning and model training to deployment and production-grade authentication — with a particular interest in solving financial inclusion problems for both the Ethiopian and international fintech markets.
Experience
AI & Machine Learning Intern
- —Built 3+ end-to-end AI and machine learning solutions for real-world Ethiopian fintech challenges, applying Python, Pandas, and scikit-learn across forecasting, credit risk modeling, and financial NLP workflows.
- —Worked across the full ML lifecycle — data cleaning, feature engineering, model training, evaluation, and explainability — on projects using real Ethiopian banking and financial complaint data.
Back End Developer
- —Designed and implemented a JWT-based authentication system supporting Buyer, Seller, and Admin roles, including token refresh handling and route-level access control.
- —Developed custom Django REST Framework APIs with role-based permissions, secure user management, and clean endpoint architecture built for production use and future scaling.
Web Development Intern
- —Built and delivered 4 responsive web applications using HTML5, CSS3, and vanilla JavaScript within a 2-month internship, focusing on cross-device consistency and clean UI structure.
Projects
Fair Fee — Ethiopian transfer fee analyzer
Fee transparency platform comparing real, published transfer fees across Ethiopian banks and mobile wallets (CBE, Awash, Dashen, Telebirr, M-Pesa). Models data with a dbt star schema and empirical fairness scoring engine served via FastAPI.
Ethiopian bank customer review intelligence
Multilingual (English, Amharic, Afaan Oromo) ML system predicting customer risk from CBE, BOA, and Dashen Bank app reviews — achieved 100% recall on high-risk reviews. Delivered SHAP-based explainability and a live interactive dashboard for non-technical stakeholders.
Credit risk probability model
Built an end-to-end ML pipeline for credit risk classification using alternative transaction data and RFM proxy variables, aligned with Basel II principles. Integrated SHAP for model interpretability and containerized the environment with Docker for reproducible, production-style deployment.
Fraud detection system
Compared Logistic Regression, Random Forest, XGBoost, and LightGBM on e-commerce and credit card transaction data; identified and fixed a cross-validation data leakage bug. Delivered SHAP-driven business recommendations, including risk scoring rules for new-account and geography-based fraud signals.
RAG-powered customer complaint chatbot
Built a Retrieval-Augmented Generation conversational agent that analyzes financial customer complaints and retrieves evidence-backed answers via semantic search, reducing response hallucinations.
SmartHire AI
Hackathon finalist project at Bahir Dar Institute of Technology (Apr 2026) — a full AI-powered hiring platform built in 48 hours.
Portfolio optimization
Applied Modern Portfolio Theory alongside ARIMA and LSTM time-series forecasting to explore quantitative investment strategy.
Technical Skills
LANGUAGES
5FRAMEWORKS
6MACHINE LEARNING & AI
9DATA & INFRASTRUCTURE
6ENGINEERING PRACTICES
6Volunteering & achievements
Startup Enabler Award
Awarded for helping Ethiopian tech startups grow through innovation and technical skills.
Hackathon Finalist — SmartHire AI
Built a full AI-powered hiring platform in 48 hours.
G17 UAC Campus Ambassador
Represent Bahir Dar University on tech & AI innovation initiatives.
Certificates & Recognition
Startup Enabler Award
Natnael Melese (NM Award)
Awarded for helping Ethiopian tech startups grow through innovation and strong technical skills.
Let's work together
Open to fintech-focused ML and backend roles, and freelance projects across credit risk, fraud, and financial NLP — for the Ethiopian and international market.