Portfolio · 2026

Open to work · Jakarta

ArkanantaRadithya

Portrait of Arkananta Radithya with arms crossed

Data & AI engineer · multimodal AI & real-time pipelines

12+

projects shipped

3.62

GPA / 4.00

80+

students mentored

About

Fig. 01 · The engineer

Data & AI at the core, engineering to ship it. I train multimodal AI that reads text and images, then turn it into products people actually use.

S.Kom graduate from Institut Teknologi Sepuluh Nopember (ITS), GPA 3.62. My focus is data and AI: multimodal models for early depression detection, real-time big data pipelines, and the analytics behind them. I also build the web products that carry those models to real users. I care about systems that work in production, not just demos.

Conference paper · First author

Presented at ICAICTA 2026 (Oct 2026): multimodal AI for early depression detection, written as my undergraduate thesis.

See achievements ↓

12+

Projects shipped

3.62

GPA / 4.00

80+

Students mentored

Experience

  1. Freelance

    Data & Machine Learning Engineer

    Aug 2026 - Present

    JokiProyek · Remote

    • •Take on client data and analysis work through JokiProyek: turn each request into a written scope of work with a delivery estimate before starting.
    • •Clean and analyse the client's dataset in Python (Pandas, Scikit-learn), then hand back the analysis, or the small model, the brief asks for.
  2. IT internship at PT Badak NGL
    Internship

    IT Intern

    Dec 2024 - Feb 2025

    PT Badak NGL · Bontang, East Kalimantan

    • •Built an RSA login flow prototype for 3 internal systems (e-Procurement, e-PNK, e-Contract) that share one login form; encrypted credentials with the public key along the login path, with database storage left to the implementation.
    • •Tested the flow with dummy data, then handed the source code to the IT team and presented and demoed it for implementation.
  3. UI/UX design for the HMIT ITS website
    Project

    UI/UX Designer

    2024

    HMIT ITS · Website Development Team · Surabaya

    • •Designed user flows and high-fidelity wireframes for the association's official web portal in Figma.
    • •Built a component-based design system (UI components, typography, color, spacing) and handed specifications to frontend engineers as the implementation reference.
  4. Teaching assistant at Institut Teknologi Sepuluh Nopember
    Teaching

    Teaching Assistant (4 Semesters)

    Feb 2023 - Jun 2025

    Institut Teknologi Sepuluh Nopember · Surabaya · 80+ students/semester

    • •Mentored 80+ students per semester across 4 core courses (Big Data & Data Lakehouse, Web Programming, Mobile Programming, Data Structures & OOP) for 4 semesters, February 2023 - June 2025.
    • •Co-authored 5 lab modules and graded 10+ final project teams per semester against code architecture, security, responsiveness, and usability standards.
    • •Taught Apache Kafka (real-time streaming) and MinIO (data lakehouse) labs and graded end-to-end data analysis final projects; held weekly consultations on programming logic and algorithms.

Index / 03 · 2025 / 2026

My
Project

Scroll to explore

Case 01 / 05

CelestX

82.10% accuracy · 79.10% F1 · 5-fold CV

Multimodal AI · Final Project · 2026

My final project: multimodal AI for early depression detection in Indonesian social media, fusing text and images.

IndoBERTweetVision TransformerPyTorchFastAPIReact.jsSHAPDocker
View project

Case 02 / 05

Big Data Movie Recommender

Kafka → MinIO → Spark → Trino · team of 5

Data Engineering · Machine Learning · Streaming · 2025

Team project (5): streaming movie recommender where real-time pipelines feed an ML model that suggests similar movies and TV shows, with a live dashboard and query layer.

KafkaMinIOSparkStreamlitTrinoPythonDocker
View project

Case 03 / 05

BoothMe AI Photo Booth

SD 1.5 revAnimated · real-time MediaPipe · team of 3

Generative AI · Web App · 2025

Team project (3): web-based AI photo booth. Snap a photo and transform it with generative AI.

Stable DiffusionVueDocker
View project

Case 04 / 05

F1 Race Strategy

366,508 laps · pit-loss rule ROC-AUC 0.52 vs 0.82 from gap + tyre age

Data Analytics · Statistical Modeling · Kaggle · 2026

Solo analytics project: tested the pit-wall rule that an undercut is on when the gap is smaller than the pit loss, across 15 Formula 1 seasons of public timing data.

PythonSQLPandasscikit-learnParquet/SQLitepytest
View project

Case 05 / 05

Retail Demand Forecasting

MAE $886 · R² 0.84 · +38.8% promo lift

Business Analytics · Forecasting · Kaggle · 2026

Solo business analytics: daily sales forecasting for 1,115 Rossmann drugstores in PySpark, with an honest temporal split and per-store target encoding.

PySparkSpark MLlibRandom ForestPandasMatplotlib
View project

Other Works

  • Kaggle notebook: an xG (expected goals) model over spatio-temporal shot coordinates, from distance and angle feature engineering to a live-match simulation sandbox.

    PySparkRandom ForestPandasMatplotlib
  • Kaggle notebook: regression on US flight records to predict arrival delays in minutes.

    PySparkRandom ForestPandasMatplotlib
  • Public dataset release: 7.9k labeled Indonesian tweets for depressive-symptom detection, LLM-augmented and clinically validated (CC BY-NC-SA 4.0).

    NLPLLaMA-3.3-70B
  • Course final project: an Android app that reads traffic signs from the camera feed and classifies them with a trained CNN.

    AndroidJavaOpenCVTensorFlow Lite
  • Team project (2): an AR shooter for Android. Point the camera around your room, find the 3D targets, and blast them.

    AndroidJavaARCoreSceneform
  • Team project (3): a location-based story sharing app. I built the main page, navigation, profile, and the story and location flows.

    AndroidJavaFirebaseCameraX
  • Contributed to the official web programming practicum module at ITS, including the responsive layout material in the HTML & CSS chapter.

    HTMLCSS

Skills

Languages & Query

Python · JavaScript · TypeScript · SQL (PostgreSQL, MySQL) · HTML/CSS · Java · C++

Data & ML

Pandas · NumPy · Scikit-learn · PyTorch · Exploratory Data Analysis (EDA) · SHAP (XAI) · Natural Language Processing (NLP) · Computer Vision · Microsoft Excel (Pivot Tables, VLOOKUP)

Engineering & Tools

Apache Kafka · Apache Spark · MinIO (Data Lakehouse) · Trino (SQL Query Engine) · Docker · Docker Compose · Git · Linux/Bash · FastAPI · Streamlit · React.js · Next.js · Vue.js (Vue 3/Vite) · Node.js (Express) · TailwindCSS · RESTful APIs · Object-Oriented Programming (OOP) · Figma (UI/UX Design) · Postman · Firebase

Achievements

Conference paper · First author

ICAICTA 2026

The 13th International Conference on Advanced Informatics: Concepts, Theory and Applications

Paper presented

My undergraduate thesis: a multimodal system that reads Indonesian social media text and images to surface early depression signals, evaluated as a first-author paper and presented at ICAICTA 2026.

Multimodal AIDepression detectionIndoBERTweetVision Transformer
Role
First author
Presented
20 October 2026
Venue
UTAR Kampar, Malaysia

Presented 20 October 2026. Proceedings publication is still pending.

Index / 06 · Recognition

First author

Certification · Language

IELTS Academic

7.0Overall band · CEFR C1
Listening 8.0Reading 8.0Writing 6.5Speaking 6.0
Test date
10 October 2026
Format
Academic · C1

Full TRF available on request.

Language credential

Oct 2026

Organization

Organization

Feb 2024 · Staff

Media & Information Staff

  • •Added roughly 3,000 followers to the department's social media across 2 management periods through publication strategy and visual communication.
  • •Co-developed the organization's brand guidelines and applied them across digital channels and print materials.

Feb 2026 · Vice Head

Vice Head of Department

  • •Led a team of 11 (8 designers and 3 coordinators) after being promoted to Vice Head of Department: creative direction, publication deadlines, and digital asset QA.

Himpunan Mahasiswa Teknologi Informasi (HMIT) ITS · Surabaya

Feb 2024 - Feb 2026

Selected designs

Fig. 01 · Campaign poster
Fig. 02 · Welcome banner
Fig. 03 · Event week design

Education

Institut Teknologi Sepuluh Nopember logo

Institut Teknologi Sepuluh Nopember

Education · Surabaya

GPA / 4.00

0.00

Data Science, Artificial Intelligence (AI/ML), and Big Data Engineering.

Academic leadership: Teaching Assistant across 4 consecutive semesters, 80+ students per semester.

Relevant coursework

Artificial Intelligence & Machine LearningBig Data & Data LakehouseCryptographyWeb ProgrammingMobile ProgrammingData Structures & AlgorithmsDatabase Systems & SQL

Progress · 8 semesters

20222026

Study log

2022 - 2026