Ali Sarabi

My Research and Projects

Here are some key projects and research work I've been involved with throughout my academic and professional career.

  • Deep Learning Framework for Time Series Forecasting

    My PhD thesis: an advanced neural network architecture for accurate long-term forecasting of high-dimensional time series, learning complex variable relationships automatically. Recipient of the Outstanding Thesis Award.

    ASU · PhD Thesis

  • Forecasting Valley Fever in Arizona

    A Graph Neural Network approach to forecasting Coccidioidomycosis (Valley Fever) across Arizona, modeling spatial and temporal disease dynamics.

    arXiv:2507.10014

  • Temporal Difference Learning for Prognostics

    A multi-task reinforcement learning framework for simultaneous Remaining Useful Life (RUL) estimation and failure classification of engineered systems.

    ASU · Research

  • ExpressReply — AI Smart Keyboard

    A privacy-first, context-aware AI keyboard for mobile that drafts replies in your voice. Led the end-to-end product and on-device AI experience at Expeditise.

    Expeditise LLC

  • ExpressExplain — RAG Knowledge Platform

    A retrieval-augmented platform with scalable backend infrastructure for document ingestion, embedding, retrieval, and citation-based AI responses.

    Expeditise LLC

  • ExCalendar — AI Desktop Calendar

    An AI-powered desktop calendar that turns natural-language intent into scheduled, organized plans across your day.

    Expeditise LLC

  • Chemical Price Forecasting

    A systematic machine learning procedure using Auto ARIMAX / XGBoost to select optimal exogenous variables and forecast world price dynamics of a chemical product across market scenarios.

    Bayer Corporation

  • Business Plan & Feasibility Studies

    Authored business plans and feasibility reports to establish Sharif Advanced Machinery Development in the Sharif Advanced Technologies Incubator.

    SAMD