I am in job market now

Milad Natanzi

PhD Candidate & ECE Researcher at WPI

Hello! I am Milad, a Ph.D. Candidate and researcher in the Department of Electrical and Computer Engineering at Worcester Polytechnic Institute (WPI).

My research lies at the intersection of O-RAN architecture, AI/ML-driven network optimization, and wireless security for 6G systems. I develop intelligent frameworks that integrate Large Language Models (LLMs) into the radio access network, focusing on intelligent control, beamforming optimization, and mitigating vulnerabilities in AI-driven traffic steering. My work includes the development of the OAIC and OAIC-T frameworks an initiative designed for testing and securing AI controllers in next-generation networks.

Beyond academia, I bring 7+ years of hands-on experience in the telecommunications industry. Previously, I served as a Data Scientist and RF Optimization Engineer at MTN Irancell and various vendors, where I focused on 4G performance analysis, KPI optimization, and revenue assurance mechanisms. I received my M.S. in Information Technology, where I worked on Software-Defined Networking (SDN) security and architecture.

5G/6G Networks Machine Learning Python O-RAN

Read My CV

Publications

News & Updates

View Timeline

Experience

2023 — Present

PhD Researcher

Worcester Polytechnic Institute (WPI) - Worcester, MA, USA

As a PhD Researcher at Worcester Polytechnic Institute (WPI), I lead research on 5G/6G wireless infrastructure, focusing on efficient spectrum management and the optimization of O-RAN architecture. I develop novel approaches by integrating Machine Learning techniques such as Continual and Reinforcement Learning into physical-layer communications tasks to achieve higher network autonomy. My work also extends to the security of next-generation systems, where I investigate LLM vulnerabilities in network control and develop robust testing frameworks like OAIC-T to ensure the reliability of AI-driven wireless ecosystems.

2017 — 2022

Data Stoward

MTN Group

During my tenure at MTN Irancell, where I progressed through three key roles as a Revenue Assurance Specialist, Data Steward, and Data Scientist, I focused on large-scale data engineering and revenue optimization. In these positions, I managed and analyzed massive volumes of Call Detail Records (CDRs) within the Ericsson Charging System, built critical monitoring infrastructure, and developed proactive control mechanisms. These initiatives enabled granular network performance tracking at the gNodeB and cell levels, directly safeguarding revenue streams for millions of subscribers.

2014 — 2017

RF Engineer

Telecommunication Vendor

Delivered RF network planning, optimization, and advanced troubleshooting across multi-vendor environments, including Huawei, Ericsson, and Nokia (3G/4G/5G). This involved spearheading KPI optimization, network performance analysis, and precise site/cell planning to eliminate bottlenecks, ensure high-quality wireless coverage, and maximize network capacity.

Skills & Tools

Programming & Data Science

Python SQL R Apache Spark Git & Version Control

Telecommunications

5G & 6G Wireless O-RAN Architecture Beamforming Optimization Software-Defined Radio (SDR) Traffic Steering

Machine Learning & AI

PyTorch / TensorFlow Reinforcement Learning (RL) Continual Learning Federated Learning Scikit-Learn

Infrastructure & Systems

Linux / Bash Docker & Kubernetes GPU Clusters Management Cloud Platforms (AWS/GCP) Apache Kafka

Selected Works

Project Menu

WeaveMind

A novel pattern-driven beamforming architecture designed for 6G environments. It utilizes Local Context Memory and Stability Anchors to dramatically improve signal reliability and targeting in high-density areas.

Python PyTorch MATLAB

AI-Driven Fuzzing

A highly scalable framework for vulnerability analysis in 5G traffic steering algorithms. It employs NSGA-II multi-objective optimization to isolate hidden protocol weaknesses before they compromise network stability.

Security Optimization

HARMONY

HARMONY brings agentic AI to O-RAN testing: a multi-agent system that autonomously manages, diagnoses, and reconfigures a live 5G SA testbed through natural language.

O-RAN Multi-Agent Systems 5G SA

FairShare

A decentralized resource sharing protocol for edge devices. Utilizes federated learning to allow edge clusters to cooperatively manage power limits without central cloud orchestrators.

Edge Computing Federated Learning