Designing robust ML systems for risk intelligence, operational analytics, and emerging market environments.
Work
Background
Engineered an end-to-end YOLOv8 object detection pipeline to automate the identification of competitor fixed broadband infrastructure from street-level imagery, replacing costly manual field surveys.
Improved overall model accuracy from 0.522 to 0.763 mAP50 (+24%) through a data-centric approach: stratified dataset splitting, physics-aware augmentation, and high-resolution training at 1248px.
Resurrected failing minority classes from near-zero (~0%) to ~50% mAP50 via targeted synthetic data injection with a strict zero-contamination validation policy.
Built a geo-spatial inference engine integrating EXIF GPS extraction and offline reverse geocoding via BPS shapefiles, transforming raw detections into regional infrastructure intelligence reports.
Credentials
DBS Foundation x Dicoding
Coding Camp Gen AI Path · 4 certs
Belajar Dasar AI
Prompt Engineering untuk Software Developer
Memulai Pemrograman dengan Python
Machine Learning untuk Pemula
Digital Talent Scholarship
Artificial Intelligence Path · 2 certs
Associate Data Scientist + Python
Data Scientist Supervisor
Fundamental of Microsoft Fabric & Azure
Capabilities
Behind the work
Muanai Khalifah Revindo
Undergraduate in Informatics Engineering at Universitas Sriwijaya, building ML systems at the intersection of technical rigor and real business problems. No grand origin story, just someone who found early on that playing with data and seeing what it reveals is genuinely fun, and then realized the interesting problems live where complexity demands explainability.
These days, I'm focused on risk intelligence and operational analytics for emerging markets, work that spans credit scoring pipelines, geospatial infrastructure mapping, and production systems that survive contact with the real world. Recent projects share the same DNA: making sense of messy data in contexts where decisions matter.
The pattern-recognition habit started with stock markets at 17, still a stock investor, still thinking probabilistically about risk. That lens shaped how I approach ML: explainability by design, not afterthought. Because when regulators ask why a loan was denied, or field teams need to act on infrastructure intelligence, a black-box model becomes a liability in regulated and operational settings.
Outside of data, I draw and care about UI/UX because the best analysis means nothing if it's trapped in unusable interfaces.
Education
Currently focused on
Organisation
Outside the terminal
Muanai Khalifah Revindo
Undergraduate in Informatics Engineering at Universitas Sriwijaya, building ML systems at the intersection of technical rigor and real business problems. No grand origin story, just someone who found early on that playing with data and seeing what it reveals is genuinely fun, and then realized the interesting problems live where complexity demands explainability.
These days, I'm focused on risk intelligence and operational analytics for emerging markets, work that spans credit scoring pipelines, geospatial infrastructure mapping, and production systems that survive contact with the real world. Recent projects share the same DNA: making sense of messy data in contexts where decisions matter.
The pattern-recognition habit started with stock markets at 17, still a stock investor, still thinking probabilistically about risk. That lens shaped how I approach ML: explainability by design, not afterthought. Because when regulators ask why a loan was denied, or field teams need to act on infrastructure intelligence, a black-box model becomes a liability in regulated and operational settings.
Outside of data, I draw and care about UI/UX because the best analysis means nothing if it's trapped in unusable interfaces.
Education
Currently focused on
Organisation
Outside the terminal