Hi, I'm Bryan Sujoso

AI Engineer & Full Stack Developer building intelligent RAG systems and scalable web applications.

Download ResumeGitHubLinkedInEmailWhatsApp

Work Experience

AI & Full Stack Developer Intern (Official Title: IT Support)

PT. Sentra Proteksi Data Teknologi Indonesia

September 2025 - June 2026
  • Architected an end-to-end Retrieval-Augmented Generation (RAG) pipeline to process unstructured government PDFs, utilizing recursive character splitting and multilingual BERT for local context understanding.
  • Engineered hybrid search logic combining traditional keyword matching with modern vector search to accurately parse informal user queries.
  • Programmed strict safety guardrails preventing AI hallucinations by constraining responses to retrieved documentation with mandatory clickable citations.
  • Developed and deployed 'Kartini.ai', transitioning it from an academic capstone into a functional corporate AI agent making complex legal, financial, and healthcare information accessible.

Operations Optimization Specialist

PT. Happy Plastikindo Utama

January 2021 - August 2022
  • Modernized operational frameworks during the COVID-19 pandemic, implementing digital ecosystems to sustain production continuity during lockdowns.
  • Consolidated fragmented paper and digital records into a centralized real-time inventory tracking dashboard.
  • Designed data-entry validation protocols using protected structures and localized interfaces, reducing manual input discrepancies by roughly 40%.
  • Automated recurrent administrative reporting pipelines via custom VBA and Python scripting, reducing task overhead from 3 hours to 40 minutes daily.

Featured Architecture

Aura.privasimu.id (Enterprise RAG System)

An end-to-end Retrieval-Augmented Generation (RAG) architecture built to process unstructured documents into accessible legal, financial, and healthcare information, featuring hybrid retrieval and strict safety guardrails.

FastAPILangChainMilvus DBReact
CodeLive Demo

Hybrid NER Academic Pipeline

Published at ICISS 2025. This containerized inference endpoint combines a fine-tuned BERT model with Regex pattern matching for automatic entity extraction and file renaming, achieving an 87% accuracy/F1-score.

Hugging FaceDockerGCP Compute Engine
Code

AI-Powered SaaS Kanban Board

A full-stack project management dashboard featuring an AI assistant that automatically parses raw meeting transcripts into structured, prioritized tickets in a PostgreSQL database.

Next.jsPostgreSQLPrismaGemini API
CodeLive Demo