AARON PATRICK NATHANIEL

Personalized recommendation systems across multiple digital products: candidate generation, ranking, content mix, evaluation, and safeguards to improve relevance, discovery, and long-term user outcomes.

Started in computer vision for fraud detection and eKYC, then moved into LLMs, RAG systems, and MLOps. Led production LLM applications, enterprise RAG solutions, and automated model deployment infrastructure.

• Production face search system on FAISS processing 10,000 to 15,000 images daily
• Led NIST facial recognition submission research; fine-tuned models on 15M images for roughly 98% 1:1 verification accuracy
• Internal LLM and RAG apps used by around 100 employees, roughly 90% accuracy, around 60% fewer hallucination and guardrail issues
• Fine-tuned custom LLMs (LoRA) matching leading provider performance for the business use case
• Compressed production CV models to under 100 MB each for fully offline on-device mobile inference
• End-to-end MLOps (DVC, Jenkins, MLflow, Docker, FastAPI): deployment time cut from 2.5 hours to around 20 minutes
• OCR and document intelligence: PPOCR, CRNN, NER in production

Bachelor in Computer Science (Data Science), Multimedia University, 2015 to 2018. CGPA 3.74