Professional Experience

Current Positions

Developer & Researcher

ForceX AI — Indonesia's first AI-for-energy platform

Building physics-informed AI agents for nuclear reactors, geothermal reservoirs, oil & gas operations, power grids, and renewable energy — designed for Southeast Asia's energy transition.

Responsibilities:

  • Architect and develop the ForceX AI platform with 12 AI products
  • Design physics-informed ML models (PINNs, GNNs, FNOs, CNN surrogates)
  • Build multi-agent orchestration systems with LangGraph
  • Deploy and maintain production infrastructure (VPS, Cloudflare, Docker)
  • Publish models and datasets on HuggingFace

Key Achievements:

  • Deployed GeoForce CNN surrogate — R²=0.997 accuracy with 57K parameters, replacing hours of TOUGH2 simulation with sub-second inference
  • Built 12 product MVPs across nuclear, geothermal, oil & gas, and renewables with 484 passing tests
  • Published model weights and geothermal reservoir dataset on HuggingFace
  • Shipped production platform at platform.forcex-ai.com
  • Built AI Testing Framework (ATP) — published on npm as @robi-atp/cli

Software Developer & AI Specialist

A Leading ICT Solutions Provider

Specializing in software products and solutions that empower media convergence in Media & Broadcast Industry, E-commerce, and other ICT-related services.

Responsibilities:

  • Architect and deploy production-grade AI systems for enterprise clients
  • Build agentic RAG systems with LangGraph for enterprise documentation
  • Design and implement MLOps pipelines with automated model training and deployment
  • Develop multimodal AI extraction layers (NLP, Vision, Audio)
  • Create comprehensive system documentation and architectural diagrams

Key Achievements:

  • Architected production-grade media platform with multimodal AI and AdCP integration (350+ pages documentation)
  • Built enterprise Agentic RAG chatbot with LangGraph self-reflection, hybrid retrieval, and Langfuse observability
  • Deployed MLOps platform with 24 trained models, genetic algorithm optimization, and Prometheus/Grafana monitoring
  • Built enterprise RAG system with 75%+ cost reduction through intelligent multi-level caching
  • Implemented multimodal AI extraction layer processing 40K+ documents with NLP, vision, and audio analysis

Teaching & Training

Training Instructor

Government Transformation Academy, BPPTIK KOMINFO, Indonesia

Course Delivered:

Fundamentals of Data Science

  • Designed and delivered comprehensive data science curriculum
  • Taught government professionals core concepts of data analysis
  • Covered Python programming for data science
  • Introduced machine learning fundamentals
  • Practical hands-on sessions with real datasets
  • Emphasized data-driven decision making

Impact:

Empowered government officials with data science skills to support digital transformation initiatives across Indonesia.

Workshops & Presentations

Webinar Speaker

Deep Learning for Medical Image Analysis - November 2024

  • Presented advanced deep learning techniques
  • Focused on medical imaging applications
  • Discussed diagnostic applications in healthcare
  • Shared insights on healthcare innovation through AI

Workshop Facilitator

FSKoM Day 2024, Matana University - October 2024

  • Presented oxygen saturation and heart rate monitoring
  • Demonstrated computer vision techniques for health monitoring
  • Hands-on demonstrations with real-time systems
  • Engaged with students and faculty

Workshop Instructor

Arduino Programming for High School Students - September 2024

  • Introduced microcontroller programming fundamentals
  • Taught Arduino basics and sensor integration
  • Conducted hands-on instrumentation projects
  • Inspired young students to explore technology

Skills & Technologies

AI & Machine Learning

  • Agentic AI: LangGraph, LangChain, multi-agent orchestration, self-reflective RAG
  • LLM Systems: RAG pipelines, hybrid retrieval (dense + BM25 + RRF), cross-encoder reranking, multi-provider fallback chains
  • Physics-Informed ML: PINNs, CNN surrogates, GNNs, FNOs, LSTM for time-series
  • Multimodal AI: NLP (spaCy, Transformers), Vision (CLIP, YOLOv8), Audio (Whisper, pyannote)
  • ML Training: PyTorch, scikit-learn, XGBoost, CatBoost, genetic algorithms

MLOps & Infrastructure

  • Deployment: Docker, Kubernetes, GitHub Actions CI/CD, model registry
  • Cloud: Google Cloud Platform (Cloud Run, Cloud SQL, Artifact Registry, Secret Manager, Compute Engine)
  • Monitoring: Prometheus, Grafana, Alertmanager, Loki, Langfuse
  • Databases: PostgreSQL + pgvector, Redis, Kafka event streaming

Development

  • Backend: Python (FastAPI), TypeScript (NestJS, Node.js)
  • Frontend: React, Next.js 14, Video.js
  • Storage: MinIO S3, HuggingFace Hub, npm registry
  • Testing: Playwright, pytest, custom AI evaluation frameworks