resume and projects

General Information

Full Name Ilham Fazri
Languages Indonesia, English

🔧 Skills

Languages Python (FastAPI, Flask, ML Frameworks & Tools), Golang (Gin, GORM, gRPC), and Java (Android Native)
Databases NoSQL (MongoDB, Firebase, Redis) and SQL (PostgreeSQL)
Technologies Virtualization(Docker), Cloud Platform(AWS), Message Broker (RabbitMQ), and CI/CD (Github/Gitlab)
Software Architecture Clean Architecture, Event‐Driven Architecture, Microservices Architecture, and Design Pattern(Creational, Structural, etc)
Area of Interest Backend Development and Machine Learning

🎓 Education

  • Aug, 2018 - Aug, 2022
    B.s. in Electronic and Instrumentation (GPA : 3.8/4.0)
    Gadjah Mada University
    • Participated in International and National Competitions in the field of electronic and software engineering.
    • Thesis: Detection and Recognition Multiple Traffic Violation Type using Video Processing Based On YOLO (Accepted in ICIC Express Lett.)
  • Oct, 2022
    Backend Curriculum
    Tokopedia DevCamp Class of 2022
    • One of 80 participants selected from 4900+ applicants} to get 7-day exclusive boot camp training for future tech leaders who have the desire to create impact through technology.
    • Learn about clean architecture, database (PostgreSQL), message queue (NSQ), caching (Redis), and monitoring (Grafana & Prometheus) in developing backend service using Golang.
  • Jan, 2021 - Jun, 2021
    Machine Learning Path
    Bangkit 2021
    • Become part of 3000 students out of 40000 applicants to take in programs created by the ministry and cultural education in collaboration with Google, Tokopedia, Gojek, Traveloka and other tech companies in Indonesia.
    • We learn basic of machine learning for computer vision, NLP and tabular data and we also learn soft skills needed to be successful in tech industries, such as team work, time management, and leadership.
    • Build Doctor.ai that can predict disease based on symptoms using machine learning as a final project.
  • 2014 - 2018
    Associate Degree in Electronics
    Cimahi State Vocational High School 1 (4 Years Program)
    • Selected to represent the West Java region in the largest national vocational competition in mobile robotic held by the Ministry of Education with the result of getting first place
    • Math National Exam Score : 100/100
    • Relevant Courses : Robotics, Digital Logic & C++

💼 Experience

  • May, 2021 - Sept, 2022
    Software Engineer, ML
    Widya Wicara
    • Developed and mantained backend API core services: speech-to-text(STT) and text-to-speech(TTS).
    • Improved the speech-to-text(STT) service for Indonesia language by reducing the WER (Word Error Rate) from 8% to 4% by implementing the newest state-of-the-art model.
    • Built a new variant (male) voice for Indonesia text-to-speech (TTS) service and developed an algorithm to improve TTS pronunciation in English words.
    • Built reliability and scalability core services using (RabbitMQ + Celery) as message broker, Flask as web framework, MongoDB for caching mechanism, Docker for virtualization, and several AWS Stack (EC2 \& S3).
  • Jun, 2021 - Jun, 2022
    Research Assistant - Freelance
    Department Computer Science and Electronics, Gadjah Mada University
    • Built a machine learning model to classifying people wearning mask and not from video in the purpose of reduce spreading of COVID-19 virus with the accuracy ~95%
    • Annotate high-resolution dataset from Borobudur and Prambanan temple relief for semantic segmentation task. The dataset has three layers, the first layer for the various object found in relief such as a human, animal, building, artifact, etc. the second layer for jewelry found in the human object, and the third layer for the human pose.
  • May 2021 - Jan 2022
    Software Engineer, ML - Intern
    Widya Wicara
    • Build a text to speech model for Indonesia language based on Deep Convolutional Networks with Guided Attention using Python and Pytorch.
    • Develop a wake-up word system for the smart speaker, uses audio Mel spectrogram as feature and uses bidirectional long short-term memory (Bi-LSTM) as classification model using TensorFlow. The model has 91% accuracy on validation data.

🎖 Honors and Awards

  • Aug, 2022
    1st Place at Paragon Corp Hackathon Competition 2022
    PT Paragon Technology and Innovation
    • Built an AI-based web app to find skin-tone colors and recommend the best color for makeup, also with a virtual try-on feature. I am Responsible as a Fullstack Developer and successfully built a web app for PoC in 24 hours based on idea creation.
  • Dec, 2020
    Awardee Puncak Persembahan Prestasi Talenta Indonesia
    The Ministry of Education, Culture, Research, and Technology, Republic of Indonesia
    • One of the students who was selected because of actively participate and got awards in national and international competitions.
  • Nov, 2020
    Double Gold Medal (Presentation and Poster) at PKM Karsa Cipta, Pekan Ilmiah Nasional (PIMNAS) ke‐33
    The Ministry of Education, Culture, Research, and Technology, Republic of Indonesia
    • Gold medal in presentation and poster in one of the biggest university competitions in Indonesia. We designed and built a device to help medical personnel detect early peripheral neuropathy symptoms to prevent amputation in diabetics. Implemented computer vision and integrated with the android application.
  • Oct, 2020
    Finalist Smart City Category, Pagelaran Mahasiswa Nasional bidang Teknologi Informasi dan Komunikasi (GemasTIK) 2020
    The Ministry of Research and Technology, Republic of Indonesia
    • Built CovaTech is an app to helped the government to reduce the spread of the COVID-19 virus.
  • Jan, 2020
    Gold Medal at Technology
    International Science Technology and Engineering Competition
    • Build a tool that can detect peripheral neuropathy early from people that have diabetes. The tool can be connected to an android application to shows data from sensors and also will give recommendations to patients based on the result. For more information, check this link
  • Oct, 2019
    2nd Place at Vertical Takef-off and Landing Division, Kontes Robot Terbang Indonesia (KRTI) 2019
    The Ministry of Research and Technology, Republic of Indonesia
    • Built and design a drone that has a mission to dropping payloads on specifics locations automatically quickly and accurately. Use python for the main controller (we used Robot Operating System)and use C++ to control the payload mechanism.
  • May, 2017
    1st Place at Mobile Robotic Category, Lomba Kompetensi Siswa (LKS) Tingkat Nasional 2017
    The Ministry of Education, Culture, Research, and Technology, Republic of Indonesia
    • Represent west java in annually national vocational student competition to design and build a robot that has a mission to avoid obstacles and to moving objects based on colors automatically using various sensors planted on the robot. Mostly, using C++ as the main programming language and implemented some popular algorithms for image processing.

🚀 Personal Projects

  • Pendek.In
    • A simple REST API for creating and customizing bio links (like linktree) with analytics features using Go. Implemented PostgreSQL as a database, Redis for caching, clean architecture (Controller -> Service -> Repository), and some twelve-factor methodology.
  • Cosmate (Paragon Hackaton Project)
    • An AI-based web app to find skin-tone colors and recommend the best color for makeup, also with a virtual try-on feature. I am Responsible as a Fullstack Developer and successfully built a web app for PoC in 24 hours based on idea creation. Implemented some machine learning algorithms, Flask for web framework and backend, bootstrap + javascript for front-end, and MongoDB for the database.
  • traffic-violation-detection
    • Traffic Violation Detection & Recognition is a system built to detect various traffic violations. Currently, this system only supports detecting running red light violations, helmet violations, and wrong-way. I used YOLOv5 for object detection and SORT as an object tracker. Implemented PyQT for GUI, machine learning frameworks (Pytorch), and various data structures and algorithms to make this possible.
  • CovaTech an android application for COVID-19 tracking.
    • A native android application that has a location tracking function that is visited and also an algorithm to provide notifications if the user has met someone who has Covid-19. Using android studio as IDE, Java as programming language, and Firebase for Authentification and Database..

📜 Certification and Courses