Fadhil Elrizanda

Master's Researcher in Artificial Intelligence

Specialized in 3D Object Detection, Multimodal Fusion, & Knowledge Distillation

Academic Research Profile

As a Master's Candidate in Artificial Intelligence at the prestigious Institut Teknologi Bandung (ITB), my academic focus is positioned at the intersection of deep learning theory and applied computer vision. My thesis research is specifically dedicated to Knowledge Distillation for 3D Object Detection, with the core objective of compressing large, highly accurate multimodal models (leveraging both LiDAR and Camera sensors) into efficient networks suitable for real-time inference on resource-constrained edge devices such as autonomous vehicles.

Selected Academic Distinctions

Gold Medal & Special Award - International IoT IYSA 2023
Silver Medal - Global Competition for Life Science IYSA 2023
Winner 2nd Place - GEMASTIK Division IX Smart Device 2022
Sakura Science Exchange Program Participant - JST Japan 2023
Outstanding Graduate - Andalas University 2023

Education & Degrees

Master of Engineering (M.Eng.) in Informatics

Specialization: Artificial Intelligence

Institut Teknologi Bandung (ITB) | Expected Graduation: Aug 2026

Thesis Research: "Optimizing Multimodal 3D Object Detection Using Camera and LiDAR Modalities via Advanced Knowledge Distillation Methodologies."

Advanced Coursework & Research Domains: Advanced Machine Learning, Probabilistic Graphical Models, Deep Neural Networks, Multimodal Sensor Fusion, and Autonomous Systems Perception.

Bachelor of Engineering (B.Eng.) in Electrical Engineering

Universitas Andalas, Padang | Dec 2023

Thesis: "Vision-based vehicle safety system using YOLOv5 for real-time detection and risk mitigation in complex environments."

AI Methodologies & Competencies

Advanced Computer Vision

Architectural design for Multimodal 3D Object Detection (LiDAR + Camera fusion). Specialized in Knowledge Distillation for compressing complex teacher models into efficient student networks.

Deep Learning Architectures

Extensive implementation experience with Convolutional Neural Networks (CNNs), Vision Transformers, and Graph-RAG (Retrieval-Augmented Generation) systems utilizing LLMs.

AI Engineering & Deployment

Proficient in Python, PyTorch, TensorFlow, and edge deployment tools like TensorRT and ONNX. Executing rigorous, reproducible experimental designs using MLOps practices.

Professional AI Experience

AI Engineer

May 2026 - Present

Riset AI

Conducting applied AI research and architecting machine learning frameworks for varied industrial and deep tech applications.

Computer Vision Engineer

Dec 2025 - Present

Autonoma

Designing sophisticated algorithms for Object Detection on Autonomous Vehicles.
Developing real-time perception pipelines integrating camera and LiDAR sensor modalities for comprehensive environmental modeling.

Capstone Project Advisor (AI)

Apr 2026 - Present

Coding Camp powered by DBS Foundation

Advising student teams on end-to-end ML system design, rigorous dataset curation, model validation, and deployment strategies.

Machine Learning Instructor

Dec 2024 - Aug 2025

Dicoding DBS

Engineered and disseminated advanced ML curricula utilizing PyTorch and TensorFlow frameworks.
Supervised applied research projects with a focus on empirical experimentation and systematic error analysis.

Research Implementations

BEVFusion Architecture for Autonomous Driving

Synthesized synchronized LiDAR and camera datasets within the CARLA environment to simulate complex autonomous driving scenarios. Trained the BEVFusion model for robust multimodal 3D object detection, accompanied by a low-latency web dashboard for inference visualization.

Multimodal Fusion 3D Object Detection

Semantic Drug Repositioning via Graph-RAG

Engineered a sophisticated biomedical QA architecture intersecting Large Language Models (LLMs) with Neo4j knowledge graphs. The framework utilizes PubMed embeddings to facilitate Cypher-based semantic retrieval, enabling advanced deductive reasoning over complex biomedical networks.

LLMs & NLP Graph-RAG

Spatial Density Estimation via CNN Regression

Architected a comprehensive monitoring methodology leveraging regression-based deep learning paradigms to infer highly accurate density maps in aquaculture environments. Bridged the gap between raw optical data acquisition and production-ready analytical models.

Density Estimation CNNs

Edge-Computing IoT Pipeline via YOLOv4

Developed a robust edge-deployment pipeline employing YOLOv4 to conduct real-time classification of broiler health states. The integrated system bridges edge inference with a centralized web dashboard, earning the Gold Medal at the 2023 IYSA International IoT competition.

Edge Inference YOLOv4

Connect & Collaborate

I am continually seeking avenues for collaboration in pioneering AI research and applied Computer Vision. Please connect with me to discuss methodologies, academic publications, or collaborative engineering opportunities.