CV

Ph.D.

Sharif University of Technology

Artificial Intelligence

M.Sc.

Sharif University of Technology

Bioinformatics

B.Sc.

University of Tehran

Computer Engineering

PhD Researcher

Rohban Lab, Sharif University of Technology

Studying how implicit inductive biases affect training dynamics and Gradient Flow, enhancing Group Robustness by mitigating Shortcut Learning and reducing reliance on Spurious Correlations.

Visiting Researcher

Vikas Lab, Aalto University

Studied the theoretical foundations of Graph Neural Networks (GNNs), exploring asynchronous message-passing mechanisms that may extend their expressivity beyond the classical 1-Weisfeiler–Lehman (1-WL) limit.

M.Sc. Researcher

Rohban Lab, Sharif University of Technology

Used the RxRx19a dataset (high-throughput five-channel fluorescent microscopy images) to estimate hit scores of drug–dose combinations against SARS-CoV-2. Extracted cellular morphological features at the single-cell level, aggregated them into treatment-level profiles, and developed a scoring algorithm reflecting antiviral efficacy.

Undergraduate Researcher

Faghih Lab, University of Tehran

Designed a parameterized synthesis technique for self-stabilizing algorithms in symmetric networks, developing tight cutoffs that guarantee closure within legitimate states and deadlock-freedom outside them.

Data Scientist

Tapsi

• Fraud Detection: Developed and maintained machine learning pipelines for automatic detection of fraudulent drivers, identifying commission evasion and fake rides generated through GPS spoofing.
• Passenger Campaign Optimization: Built and deployed tools to optimize passenger marketing campaigns, maximizing return on investment (ROI) as a key performance indicator (KPI).

Bioinformatician, supervised by Prof. Sharifi Zarchi

Pardis Gene

• Rare Disease Genomics: Designed and implemented algorithms to identify causal genetic variations (SNVs, CNVs, and structural variants) using parent–child whole genome and exome sequencing data integrated with clinical phenotypes and physician diagnoses.
• Precision Oncology: Developed automated pipelines for detecting somatic variants (SNPs, indels, CNAs, fusions, and MSI) from paired tumor–normal sequencing data, and generated personalized drug recommendations based on actionable alterations.

RISE-MICCAI 2023 Travel Award

Fully funded travel award (USD 3,000).

Honorable Mention

ICPC Asia Regional Contest

On the Role of Implicit Regularization of Stochastic Gradient Descent in Group Robustness

International Conference on Learning Representations (ICLR)

The Silent Helper: How Implicit Regularization Enhances Group Robustness

High-dimensional Learning Dynamics (ICML Workshop)

Snuffy: Efficient Whole Slide Image Classifier

ECCV 2024, Springer LNCS

Navigating the MIL Trade-Off: Flexible Pooling for Whole Slide Image Classification

NeurIPS 2025

Weakly-Supervised Drug Efficiency Estimation with Confidence Score: Application to COVID-19 Drug Discovery

MICCAI 2023, Springer LNCS 14244

MILFORMER: Weighted Dual Stream Class Centered Random Attention Multiple Instance Learning for Whole Slide Image Classification

AI for Health Equity and Fairness 2024, Springer Nature

Parameterized Synthesis of Self-Stabilizing Protocols in Symmetric Networks

Acta Informatica

Parameterized Synthesis of Self-Stabilizing Protocols in Symmetric Rings

OPODIS 2019

ICLR Workshop Co-organizer

Spurious Correlation and Shortcut Learning: Foundations and Solutions Workshop.

Co-instructor, Bioinformatics Algorithms

Sharif University of Technology