My Engineering Philosophy: "Don't just write code that works. Break down the system, optimize the underlying resource layouts, and build data workflows that scale structurally."

• 🔭 I’m currently working on "Sekolah Rakyat" (Indonesian Education Expert LLMs) at the Artificial Intelligence Talent Factory (AITF) and researching at the Matsuo Iwasawa Lab, University of Tokyo (GCI Global) .
• 🌱 I’m currently learning advanced data analytics and machine learning methodologies to optimize distributed cloud architectures .
• 👯 I’m looking to collaborate on open-source AI engineering, high-throughput ETL pipelines, and scalable LLM infrastructure .
• 🤔 I’m looking for help with advanced multi-node GPU cluster orchestration and distributed vector databases.
• 💬 Ask me about Python development, Apache Airflow pipelines, RAG systems, and LLM preference alignment (SFT/DPO/GRPO) .
• 📫 How to reach me: Connect via LinkedIn or visit my portfolio at lalalostcode.dev .
• 😄 Pronouns: He / Him
• ⚡ Fun fact: I am an active Java evader who deeply prefers building machine learning workflows from the ground up using Python .

• High-Throughput Ingestion: Designing asynchronous scraping architectures and parsing pipelines to extract and process massive unstructured data streams cleanly. * Model Customization & Quantization: Fine-tuning open-source language models utilizing accelerated compute kernels (QLoRA) and preference alignment protocols (SFT/DPO/GRPO). * Containerized Infrastructure: Orchestrating multi-component environments using Docker, Kubernetes, and automated workflow DAGs for reproducible states. ---

I am always open to discussing data pipeline optimizations, distributed system designs, or open-source AI infrastructure. * Professional Network: LinkedIn * Technical Docs: lalalostcode.dev ```