Portfolio

Math Graduate. Data and ML. Occasional Game Dev.

Alief Randiansyah
Pradanaputra.

Hi! I'm Alief, a mathematics graduate from ITS who loves turning messy data into things people can actually use. Most of my projects start the same way, a website full of useful information and no easy way to get it. So I build the scraper, train the model, and sometimes wrap the whole thing in an app you can talk to.

Portrait of Alief Randiansyah Pradanaputra

Projects

★ Self-Retraining-Model-Platform ↗

MLOps platform running one closed loop. Drift detection triggers automatic retraining, conditional promotion, and zero redeploy serving.

More about this project

One FastAPI app over real infrastructure, Postgres, MinIO, MLflow, and Argo Workflows on Kubernetes. An 8 node DAG runs ingest, validate, feature engineering, Optuna search next to a PyTorch MLP branch, evaluation with a conditional promotion gate, then SHAP and promotion. A cron workflow batch scores every hour, and a React dashboard watches the loop live.

If I revisited it, I would use stronger drift statistics, MLflow model aliases instead of pinned stages, and deeper feature engineering.

  • Python
  • FastAPI
  • Postgres
  • MinIO
  • MLflow
  • Argo Workflows
  • Kubernetes
  • DVC
  • Optuna
  • PyTorch
  • React
  • Docker
  • MLOps
  • Drift Detection

★ Lights-vs-Numbers ↗

Data analysis: Indonesia's official GDP against independent activity signals, PortWatch shipping, search trends and night lights. 42-quarter sourced dataset, annotated Plotly charts, Streamlit dashboard plus a live static mirror, CI green.

  • Python
  • pandas
  • Plotly
  • Streamlit
  • GitHub Actions

QUACKO ↗

QA suite for an OCR receipt reader. A Flask plus Tesseract service covered by pytest and Playwright suites (API contract, degradation, regression, UI), CI on every PR and push, reports on an evidence site.

  • Python
  • Flask
  • Tesseract
  • OpenCV
  • pytest
  • Playwright
  • GitHub Actions

tanya-sekolah ↗

Natural language Q&A over 2,269 Surabaya and Sidoarjo schools, with tuition estimates for the schools that never publish fees. Hand written LLM routing, text to SQL guard, and RAG.

  • Python
  • LLM
  • RAG
  • Question Answering
  • SQLite
  • ChromaDB
  • fastembed
  • CatBoost
  • OpenAI compatible SDK
  • Selenium
  • Docker

InternProject-SPP_Sekolah ↗

SEVIMA intern project. Scraping notebooks and grid search cross validation behind the school tuition model. Four government portals unified into one dataset.

  • Python
  • Selenium
  • Jupyter
  • Cross validation
  • Web Scraping
  • Machine Learning
  • CatBoost

KacaSaham ↗

Weekly Telegram digest that scores IDX stocks 0 to 100 on trust and resilience, with an evidence link for every claim. Sectors Hackathon 2026.

  • Python
  • SQLite
  • Telegram
  • Telegram Bot API
  • Sectors API
  • Stocks
  • Data Analysis
  • LLM

DAWG-ID ↗

Portal visualizing a macro shock vulnerability index for 38 Indonesian provinces, with choropleth maps and the index dataset. Top 20 entry in the Microsoft AI Impact Challenge Datathon.

  • Python
  • React
  • Vite
  • JavaScript
  • GeoJSON
  • Geospatial
  • Data Visualization
  • Jupyter

Web scrapers!

Many of them!

Dapodikdasmen, Sekolah Kita, and Verval Yayasan are school portals, and not one of them offers an API. So each gets its own scraper, built around how the site actually behaves. Dapodikdasmen gives up school listings district by district with retry logic and batch CSV, Sekolah Kita profiles come down by NPSN code with checkpoint and resume, and Verval Yayasan gets crawled six levels deep with async Python. Together they feed one dataset, which turned into InternProject-SPP_Sekolah, the tuition prediction project, and then tanya-sekolah, the LLM assistant built on top of it.

PDDIKTI is educational too but serves a different purpose, covering universities rather than schools in the national higher education database. Scraped across semesters with a multithreaded Selenium setup.

Separate from all of that sits LinkedIn jobs, scraped with login persistence, random delays, and anti-detection.

Experience

Data Analyst. SEVIMA, Surabaya

Feb 2025 to Aug 2025

Analyzed active user thresholds and scored prospective K-12 customers with machine learning. Scraped school and PDDIKTI data for SEVIMA and its clients.

Team Leader, ITS Representative. Satria Data, Big Data Challenge

Aug 2025 to Sep 2025

Led the ITS team in the Big Data Challenge division after passing the internal selection as a solo programmer.

Laboratory Assistant. Programming and Visual Computing Lab, ITS

Aug 2024 to Dec 2024

Assistant for the Algorithms and Programming practicum, teaching first year students.

Game Developer, IT Division. OMITS 17

May 2024 to Oct 2024

Built a maze themed math olympiad game in Godot Engine.

Programmer. Ganesha Interactive Media (GIM), Bandung

Sep 2023 to present

Game programming practice in Godot Engine in a multi role team.

Education

Institut Teknologi Sepuluh Nopember (ITS). Bachelor of Mathematics

Aug 2022 to Aug 2026, Surabaya

Certifications

Mathematical Data Analytics Bootcamp. Matematika ITS

Sep 2026

Getting Started with Docker. Simplilearn

Sep 2026

Python for Data Science and Machine Learning Bootcamp. Udemy

Nov 2024

Intro to Machine Learning. Kaggle

May 2023

Scientific Computing with Python. freeCodeCamp

Jul 2021

Honors and awards

Top 20, AI Impact Challenge Datathon. Microsoft

Jun 2026

Awarded for DAWG-ID, the macro shock vulnerability portal for 38 Indonesian provinces.

Skills

Contact