Rifa Naftali Azka

AI/Machine Learning Engineer and Data Analyst

Aspiring AI/ML Engineer focused on developing practical machine learning solutions and intelligent applications. Experienced in exploring Computer Vision, Natural Language Processing, and data-driven approaches through hands-on projects. Passionate about transforming complex problems into scalable, meaningful solutions.

Portrait of Rifa Naftali Azka wearing glasses and a dark blazer
Jakarta, IDOpen to Internships

About

I’m a Computer Science student majoring in Artificial Intelligence at BINUS University (2024 to 2028). I like using machine learning and data analysis to solve practical problems, from detecting child stunting risk to checking produce freshness from a photo.

Outside code, I’ve worked in video production, design, mentoring, and direct client communication. That background helps me explain technical work clearly and build things people can actually use.

Recent highlights: BINUS representative at GEMASTIK XIX 2026 (Software Development), 8th place in the Web Design Competition 2026 HIMA RPL Polindra, and AMD ROCm Certified Associate.

Projects

Noise-Robust TB Cough Screening with HuBERT

2026 · Research project, BINUS

Slide: patient-level AUROC of the HuBERT model across noise levels

A research project on how background noise affects tuberculosis cough classification. Frozen HuBERT embeddings feed a patient-level classifier that I tested across six noise levels and compared with MFCC + SVM and Random Forest baselines.

Evaluation used patient-level cross-validation with bootstrap confidence intervals (121 patients, about 10,000 cough clips), a label-permutation check (mean AUROC 0.49), and an audit of cohort-label confounding. On clean audio the model reached an AUROC of 0.71; noise augmentation lowered that slightly (0.68) but kept far more sensitivity in heavy noise. These are exploratory results, not clinical claims.

Python, PyTorch, torchaudio, HuBERT, scikit-learn, Google Colab

GPU Computing with AMD ROCm

2026 · ROCm Certification labs

Bar chart of measured speed-ups from the ROCm optimization lab

Hands-on labs from the AMD ROCm Certification Program. I benchmarked rocBLAS matrix multiplication (about 67,000 GFLOPS, 41% of FP32 peak), MIOpen convolution autotuning (5,237 ms first run, 3.0 ms cached), and rocFFT round trips, trained ResNet-18 on CIFAR-10 with PyTorch on ROCm, and debugged a GPU visibility misconfiguration.

In the optimization lab, a shared-memory transpose, a workgroup-size sweep, and FP16/BF16 mixed precision gave measured gains of about 1.1x, 1.2x, and 3.6x to 3.7x.

ROCm, HIP, rocBLAS, MIOpen, rocFFT, PyTorch

AI-Powered Produce Intelligence

2025 · Multimodal AI

Produce Intelligence: chat interface, architecture diagram, diagnostic cards and model metrics

A platform that gives real-time freshness scores from images and text. I combined ResNet50 with a Phi-2 model fine-tuned with 4-bit QLoRA, deployed it on Google Cloud Platform with Firebase and OAuth, and built the chat interface and metrics dashboard.

Python, PyTorch, QLoRA, LangChain, RAG, Google Cloud Platform, Firebase

Stunting Care

2026 · GEMASTIK XIX

Stunting Care website screens

A growth-tracking and nutritional risk system that helps Posyandu and Puskesmas health workers spot stunting early. I built the PyTorch risk model, the React dashboard, and the UI design. Ranked 8th in the Web Design Competition 2026 HIMA RPL Polindra.

PyTorch, FastAPI, React, Tailwind CSS

Agentic AI Chatbot for Stunting

2025 · MLOps course

Stunting chatbot app: chat, info, news and author pages

A web app that answers stunting and nutrition questions using an LLM with RAG. I built the retrieval pipeline (sentence-transformers and Chroma DB), the Flask API, and the web interface.

Flask, Chroma DB, RAG, Docker

PPE Detection

2024 · Deep Learning

PPE detection web app with live camera and upload panels

A YOLOv8 web app that checks for hardhats, safety vests, and masks from a webcam or uploaded image, counts people, and speaks a warning when equipment is missing.

YOLOv8, OpenCV, Flask, PyTorch

Rice Variety Classifier

2026 · Computer Vision

One sample grain image from each of the five rice varieties

ResNet18 transfer learning with a custom classifier head to tell five rice varieties apart. It reached 98.9% accuracy on a 2,450-image test set.

PyTorch, torchvision, Matplotlib

Cyberbullying Tweet Classifier

2026 · NLP

Bar chart of F1 score per cyberbullying category

A prompt-engineered Qwen2.5 7B model running locally through Ollama, classifying tweets into six categories. I evaluated it with confusion matrices and per-class F1 to see where prompt-only classification falls short.

Ollama, Qwen2.5, scikit-learn, pandas

Earthquake Tsunami Prediction

2024 · MLOps

Earthquake tsunami data visualisation dashboard

An end-to-end pipeline on 782 earthquakes. A tuned Random Forest reached about 93.6% accuracy across 22 MLflow-tracked runs, packaged with Docker and conda.

scikit-learn, MLflow, Docker, Plotly

LibrAI

2026 · AI hackathon

LibrAI book recommendation app user flow

An AI book discovery app that recommends books from reader mood, topics, and preferences. I designed the user journey and refined the front end.

Flask, Node.js, Supabase, Qwen2.5, Figma

Mindy

2025 · Human-Computer Interaction

Mindy app user journey from onboarding to mood statistics

A mindfulness app for tracking daily emotions, with an AI companion and personal mood statistics. I designed the flows, built the screens, and wrote the backend endpoints for profiles, journal history, and sign-in.

Figma, Firebase, SQL, OAuth, JavaScript

Nest

2026 · UI/UX design

Nest smart city app screens and lean canvas

A smart-city app concept that puts public transport routes, waste bin locations, and green spaces in one place. I designed the eight screens in Figma, the visual style, and the lean canvas.

Figma, Canva

FontDale Luxury Mall

2025 · UI/UX design

FontDale luxury mall website pages

A five-page website design for a high-end mall: store directory with category filters, events, newsletter sign-up, a navy and gold palette, and a 16-icon set.

Figma, HTML, CSS, JavaScript

Skills

Tap a skill to see the projects that use it.

AI and machine learning
Python, PyTorch, scikit-learn, ResNet, YOLOv8, OpenCV, QLoRA fine-tuning, RAG, LangChain, Chroma DB, Ollama, prompt engineering
GPU computing and acceleration
ROCm, HIP, rocBLAS, MIOpen, rocFFT
MLOps and cloud
MLflow, Docker, Google Cloud Platform, Firebase, OAuth, Git
Data analysis
pandas, SQL, Matplotlib, Seaborn, Plotly, Power BI, Excel
Web and APIs
Flask, FastAPI, Node.js, React, Tailwind CSS, REST APIs
Design
Figma, Adobe XD, UI/UX prototyping

Experience

  • Freelance Videographer and Video Editor
    Bicara IPTEK Podcast, Jakarta
    Mar 2026 to present
  • Mentor and trainer, Led Learning and Empowering
    BINUS University · 5 training sessions, 8 freshmen mentored, and co-led public speaking and confidence workshops at a local school
    Mar 2026 to Jun 2026
  • Graphic Design and Events
    Nippon Club Organization, BINUS University
    Dec 2024 to Nov 2025
  • Social Media Marketing Intern
    PT Ken Yukagen Indonesia, Karawang
    Apr 2021 to Jul 2021

Credentials

  • AMD ROCm Certified Associate
    AMD AI Academy
    Oct 2026
  • Real-World AI with Claude
    Coursera and AWIT, sponsored by Anthropic
    Jul 2026
  • Generative AI courses
    Vanderbilt University via Coursera
    2026
  • Data Fundamentals
    IBM SkillsBuild
    Oct 2026
  • Hackathon JOINTS 2026, participant
    FMIPA UGM, student category
    Nov 2026
  • IELTS Academic 5.5 (CEFR B2)
    English · also German A1
    Feb 2024

Certificates

Click a certificate to enlarge it.

Certificate: AMD ROCm Certified Associate
AMD ROCm Certified Associate
AMD AI Academy
Certificate: Real-World AI with Claude
Real-World AI with Claude
Coursera and AWIT
Certificate: Generative AI course
Generative AI course
Vanderbilt University
Certificate: Generative AI course
Generative AI course
Vanderbilt University
Certificate: Generative AI course
Generative AI course
Vanderbilt University
Certificate: Data Fundamentals
Data Fundamentals
IBM SkillsBuild
Certificate: UX Design course
UX Design course
Google
Certificate: UX Design course
UX Design course
Google
Certificate: Hackathon JOINTS 2026, participant
Hackathon JOINTS 2026, participant
FMIPA UGM
Certificate: Code The Future: AI Hackathon 2026
Code The Future: AI Hackathon 2026
GenDigital Academy
Certificate: Web Design Competition 2026, participant
Web Design Competition 2026, participant
HIMA RPL TechSoft
Certificate: Applied NLP seminar
Applied NLP seminar
NUNI and BINUS
Certificate: IoT and ML simulation seminar
IoT and ML simulation seminar
NUNI and BINUS
Certificate: Market Research and Business Communication (C2.2)
Market Research and Business Communication (C2.2)
BINUS Beelingua
Certificate: Advanced English (C2.1)
Advanced English (C2.1)
BINUS Beelingua
Certificate: Professional Office (C1.2)
Professional Office (C1.2)
BINUS Beelingua
Certificate: Creative Writing (C1.1)
Creative Writing (C1.1)
BINUS Beelingua
Certificate: English Independent User (B2.2)
English Independent User (B2.2)
BINUS Beelingua

Let’s talk

I’m looking for an AI/ML Engineer or Data Analyst internship, February 2027 to January 2028, remote or onsite in Jakarta.

rifanazka02@gmail.com

LinkedInGitHub
Email copied