Hi, I'm Irsyad 👋
I’m an Indonesia-based AI Engineer with a strong focus on computer vision, specializing in machine learning and deep learning
Syd

About

As an AI Engineer based in Indonesia, I have a deep passion for computer vision and specialize in building innovative solutions using machine learning and deep learning. I am committed to automating business processes and exploring new challenges in the AI field. With a background in software engineering, I am always driven to leverage cutting-edge technologies to create impactful change.

Work Experience

E

ENB Mobile Care

March 2025 - Present
AI Engineer
Handling the Computer Vision project 'Nielai' by building a robust end-to-end ML pipeline on GCP Vertex AI using DVC, MLflow, and GitLab CI/CD. Deployed models through Flask APIs and implemented monitoring with Grafana. Designed and developed an AI-powered chatbot for Trade-in services, enabling users to check device prices, schedule pickups, and seamlessly connect with live agents. Worked closely with cross-functional teams to deliver scalable and production-ready AI solutions.
L

Lawencon Internasional

April 2024 - December 2024
AI Engineer
Led AI initiatives that improved recruitment workflows, including a CV parsing and scoring engine (cutting manual review by 50%) and a live coding assessment platform (reduced test time by 30%). Developed AI chatbots with 1,000+ monthly interactions and a virtual assistant for HRIS task automation. Increased candidate shortlisting accuracy by 25%.
I

Indonesia AI

August 2023 - February 2024
Junior AI Engineer
Engineered AI systems for real-world applications, including a vehicle detection system with 86% accuracy and an AI-based waste sorting system achieving 82% accuracy. Contributed to environmental AI initiatives and collaborated with cross-functional teams to implement scalable solutions.
I

Indonesia AI

February 2023 - August 2023
Computer Vision Bootcamp Batch 1
Completed intensive bootcamp focusing on real-world computer vision projects: cardiac MRI segmentation with 2D UNet, face recognition using CelebA dataset, real-time person tracking via COCO dataset, and self-driving car simulation using advanced CV techniques.

Education

T

Telkom University

2019 - 2023
Bachelor's Degree of Computer Science

Tech Stack

programming

Python
HTML

frameworks

MLflow
PyTorch
TensorFlow
TorchVision
TorchAudio
ONNX
ONNX Runtime
OpenCV
Scikit-Learn
Flask
FastAPI
Django
NumPy
Pandas

databases

MongoDB
PostgreSQL
MySQL
DBeaver

tools

Label Studio
Roboflow
Git
GitLab
Docker
Postman
NVIDIA
Jupyter
Anaconda
Visual Studio Code

cloud

Amazon AWS
Google Cloud
My Projects

Check out my latest work

I've led a range of AI initiatives—from rapid proof-of-concept models to production-scale machine-learning systems such as computer-vision pipelines and conversational agents. Here are a few highlights.

AI-Recruitment

Pipeline rekrutmen berbasis NLP & RAG: embedding OpenAI, vector store Pinecone, dan servis inference FastAPI. Dirancang untuk mempercepat penyaringan kandidat dan analitik biaya GPT.

Python
FastAPI
OpenAI
Pinecone
Docker
MLflow
Source

YOLOv11 Detection API

FastAPI + Docker untuk serving YOLOv11l. Pengguna dapat meng-upload gambar, memperoleh hasil deteksi, dan men-download gambar ber-annotasi melalui static file server.

FastAPI
YOLOv11l
Docker
Uvicorn
Python 3.11
Source

Person-Detection (Faster RCNN + COCO)

Eksperimen Faster RCNN–GoogleNet untuk mendeteksi orang pada dataset COCO (train 3 000 / test 500) menggunakan FiftyOne. Mencapai mAP 0.245.

PyTorch
FiftyOne
COCO
GoogleNet
Jupyter Notebook
Source

Cityscapes UNet Segmentation

Implementasi UNet pada dataset Cityscapes: Adam optimizer, LR 1e-3, epoch 100, augmentasi resize-256. Mendapatkan Dice Coef 0.912.

PyTorch
UNet
Cityscapes
OpenCV
NumPy
Source
Contact

Get in Touch

Want to chat? Just shoot me a DM on Twitter and I'll respond whenever I can. I ignore all soliciting.

LinkedIn
X
youtube