Summary

AI/ML Engineer with a strong background in Machine Learning and MLOps across various domains including information retrieval (Search, Recommendations, RAG-systems & ChatBots), NLP, and AI Agents.

Work Experience

• AI/ML Engineer (NLP Research Intern)

  2025-08 - 2026-02
  Trivago

  Central ML team, unstructured data systems.

  • Built semantic search engines using embedding models, improving query-to-accommodation matching accuracy across millions of listings.
  • Applied Generative AI to automate accommodation description generation and Computer Vision for automated image quality scoring and optimisation at scale.
  • Designed end-to-end ETL pipelines for unstructured data ingestion powering downstream ML systems.
• Machine Learning Researcher

  2023-09 - 2025-05
  University of Turin (DeepHealth Project)

  Federated clinical AI; industrial partners: Philips, Thales.

  • Assisted in designing federated learning workflows enabling decentralised model training across hospital networks without sharing raw patient data.
  • Participated in technical discussions aligning clinical requirements with HPC infrastructure for two major industrial partners.
  • Contributed to implementing GDPR-compliant differential-privacy algorithms ensuring data sovereignty within strict hospital IT environments.
• Data Scientist

  2021-03 - 2023-07
  Snapp!

  Ride-hailing platform with ~1M daily trips.

  • Supply-demand forecasting: Developed ML models to balance driver supply and passenger demand, materially reducing average wait times.
  • Dynamic pricing: Designed real-time surge-pricing algorithms to maximise revenue during high-demand periods.
  • ETA prediction: Leveraged streaming data to improve arrival-time accuracy for millions of daily trips.
  • Analytics: Monitored operational KPIs and executed data-driven optimisations, improving driver completion rates.
• Data Scientist

  2019-05 - 2021-03
  Digikala Group

  Largest e-commerce platform in MENA.

  • Recommendations: Engineered hybrid engine (Association Rules + Collaborative Filtering) to redesign 'Frequently Bought Together,' boosting Average Order Value via cross-selling.
  • Flash-sale pricing: Automated candidate selection and discount depth for 'Shegeftane' flash sales, achieving high sell-through without eroding margins.
  • Demand forecasting: XGBoost/Prophet pipelines capable of handling Black Friday traffic loads, reducing logistics bottlenecks and improving inventory accuracy.
  • Search relevance: Integrated custom Persian NLP models for semantic matching and misspelling correction, reducing zero-result queries significantly.
• Project Contributor

  2018-09 - 2019-03
  Sharif University of Technology

  • Institutional Analytics: Created robust analytical dashboards to support institutional decision-making and enhance strategic academic planning.

Education

• Master of Science in Language Technology & Digital Humanities (NLP)

  2026-06
  University of Turin

  GPA: 28.8/30

  Courses: Mechanistic interpretability, RAG systems, Multilingual NLP
• Bachelor of Science in Computer Science

  2019-03
  Sharif University of Technology

  Courses: Algorithms, Data structures, Statistical learning, Distributed systems

Skills

Publications

• Identity, Toxicity, or Complexity? A Language-Specific Feature Selection Approach to Reclamatory Intent Detection

  2026
  Proceedings of EVALITA 2026 (Task A: MultiPRIDE), Bari, Italy

  1st Place (Italian Task): Developed a Hybrid Fusion architecture combining BERT embeddings with engineered sociolinguistic features to detect reclaimed slurs; achieved SOTA F1 of 0.8981 by modeling language-specific syntactic patterns.
• Parametric Stubbornness: Mechanistically Isolating the Layer Shift and Sparsity Gradient of RAG Knowledge Conflicts in Llama-3

  2026
  ACL Student Research Workshop 2026

  Two-phase activation patching on Meta-Llama-3-8B across 452 minimal-pair conflicts; introduces the Sparsity Gradient and Contextual Contamination phenomena in RAG knowledge conflict settings.

Languages

• English

  Full professional proficiency
• Persian

  Native
• German

  Limited working proficiency
• Italian

  Elementary proficiency