Software Engineer @ RantAI · Jakarta, Indonesia

I build agentic AI systems that have to survive contact with production. Rust and TypeScript, mostly.

Two codebases take up most of my time:

RantAIClaw — a multi-agent runtime for production AI employees, written in Rust. I built its knowledge base (spanning a TUI and a Next.js client), multi-turn chat with channels over an SSE gateway, and the session-auth gate. I also package the runtime as self-hostable Docker images, with CI that runs security audits and fuzz tests on every push.

RantAI Agents — an agentic AI platform in TypeScript. My main piece is the RAG knowledge base: a 12-format ingest pipeline with hybrid retrieval over SurrealDB, RRF fusion, and Cohere/vLLM rerankers. Earlier I built the visual workflow engine — 20+ nodes, a custom compiler, durable state — and the real-time chat engine on Vercel AI SDK v6.

I came into this from machine learning (TensorFlow, Bangkit Academy, a Computer Engineering degree at Telkom University), and spent two years as a lab assistant teaching Python and C practicums. Still the fastest way I know to find out whether you actually understand something.

Languages

Frontend

AI / LLM

Backend

Data & Storage

DevOps & Tooling