luthfi = { "name": "Luthfi Mirza Darsono", "role": ["Software Engineer", "Full-Stack Developer"], "workplace": "Gunadarma University Computing Center (GUCC)", "position": ["Front-End Staff","ML Enginer"], "education": "Gunadarma University โ Depok, Jawa Barat ๐ฎ๐ฉ", "focus": ["Recommendation Systems", "Web Application Development", "NLP & Speech AI", "ML Pipeline Design"], "currently": "Building scalable AI-powered apps & clean UIs", }
| Project | Description | Tech Stack |
|---|---|---|
| ๐ต noice-audio-recommendation-system | Hybrid audio recommendation engine โ content metadata + synthetic listening events. Evaluated with Precision@K, Recall@K & nDCG | Python Pandas FastAPI scikit-learn |
| ๐ค AI-Interview-Evaluation-System | End-to-end AI interview analysis โ Whisper ASR, NLP scoring, PostgreSQL backend + Next.js dashboard & Streamlit | Python FastAPI Whisper Next.js PostgreSQL |
| ๐ tradescope | Crypto futures signal dashboard with technical analysis, rule-based signal scoring & AI-assisted market review | JavaScript |
| ๐ฐ Implementasi-Analisis-Sentimen-BeritaSaham | Laravel-based IDX stock sentiment analysis dashboard โ news aggregation, technical ranking, prediction evaluation, DSS backtesting & trade journal. Undergraduate thesis project | Laravel PHP NLP MySQL |
| Area | Details |
|---|---|
| ๐ฏ Recommendation Systems | Hybrid (Collaborative Filtering + Content-Based), cold-start handling, catalog-scale data |
| ๐ฃ๏ธ NLP & Speech Processing | Whisper ASR, NLP scoring, sentiment analysis |
| โ๏ธ Model Serving | FastAPI, RESTful API design, end-to-end ML pipeline |
| ๐ Evaluation Metrics | Precision@K, Recall@K, nDCG, RMSE, A/B testing mindset |
| ๐ง Data Engineering | Pandas, SQL, synthetic data generation, feature engineering |
