Sepehr Asgarian

AI Engineer at Chubb

About Me

Sepehr Asgarian

I am an Applied Scientist specializing in Large Language Models, retrieval systems, and multimodal AI. Currently, I work as an AI Engineer at Chubb, designing agentic LLM systems for insurance workflows. Previously at Klick Inc., I built production AI systems including recommendation engines, search platforms, and multimodal frameworks for healthcare applications. My work has been published in AAAI, NeurIPS, ACM Multimedia, and IEEE Transactions.

I earned my MSc in Computer Science at Western University under Dr. Boyu Wang and Dr. Yalda Mohsenzadeh, focusing on deep learning for brain-computer interfaces. I hold a BSc in Computer Engineering from Amirkabir University of Technology. My research spans generative AI, contrastive learning, and neural signal processing with applications in healthcare, multimedia, and enterprise systems.

Selected Publications

SynthAgent Paper

SynthAgent: A Multi-Agent LLM Framework for Realistic Patient Simulation

Arman Aghaee, Sepehr Asgarian, Jouhyun Jeon

In AAAI 2026 Workshop on Health Intelligence, 2026

ABS
Layer-Informed Paper

Layer-Informed Memorability Prediction and Reinforcement-Guided Multimedia Content Adjustment

Sepehr Asgarian, Jouhyun Jeon

In ACM Multimedia 2025 on Multimedia Content Generation and Evaluation, 2025

ABS
MindMem Paper

MindMem: Multimodal for Predicting Advertisement Memorability Using LLMs and Deep Learning

Sepehr Asgarian, Qayam Jetha, Jouhyun Jeon

In AAAI 2025 on Economics of Modern ML: Markets, Incentives, and Generative AI, 2025

ABS
Multiview Paper

Multi-view Contrastive Learning for Unsupervised Domain Adaptation in Brain-Computer Interfaces

Sepehr Asgarian, Ze Wang, Feng Wan, Chi Man Wong, Feng Liu, Yalda Mohsenzadeh, and 2 more authors

In IEEE Transactions on Instrumentation and Measurement, 2024

ABS
LOVENet Paper

All You Need Is LOVE: Large Optimized Vector Embeddings Network for Drug Repurposing

Sepehr Asgarian, Sina Akbarian, Jouhyun Jeon

In NeurIPS 2023 on New Frontiers of AI for Drug Discovery and Development, Dec 2023

ABS
Stock Prediction Paper

Stock Prediction Using Generative Adversarial Network (GAN) & Sentiment Analysis

Sepehr Asgarian, Rouzbeh Ghasemi, Saeedeh Momtazi

In Concurrency and Computation: Practice and Experience, 2022

ABS

Featured Projects

TradeFlare.ai Platform

TradeFlare.ai

An automated multi-agent AI platform for stock market analysis. TradeFlare screens 900+ NASDAQ stocks daily using ML filters and orchestrates 15 specialized LLM agents (Technical, Fundamental, Sentiment, Risk) to generate actionable BUY/SELL/HOLD signals. The platform delivers daily recommendations via email and maintains comprehensive track records for transparent performance evaluation.

Visit Platform

Education

Western University

2021 - 2023

MSc Computer Science, Specialization in Machine Learning

Focused on advanced machine learning, deep learning, computer vision, and statistical learning theory. Thesis research involved developing novel neural network architectures for brain-computer interface applications under the supervision of Dr. Boyu Wang and Professor Yalda Mohsenzadeh. CGPA: 4.0/4.0

Amirkabir University of Technology

2016 - 2021

BSc Computer Engineering

Comprehensive foundation in computer engineering including algorithms, data structures, software engineering, database systems, and machine learning fundamentals. Specialized projects in deep learning applications for anomaly detection and natural language processing. CGPA: 3.56/4.0

A Little More About Me

Beyond my passion for Machine Learning and AI research, I'm an avid guitarist who has been playing for nearly 12 years. Music serves as my creative outlet and helps me think differently about problem-solving. Whether I'm working through a complex algorithm or learning a new song, I find that both coding and guitar playing require patience, practice, and attention to detail.

When I'm not immersed in code or research papers, you'll often find me with my guitar, exploring different genres and techniques. I believe that having diverse interests outside of work not only keeps life interesting but also brings fresh perspectives to my professional endeavors.