Biography

I am a Ph.D. student in Computer Science at the University of Massachusetts Amherst. My research is advised by Prof. Mohammad Hajiesmaili and Prof. Ramesh K. Sitaraman. My work focuses on designing robust learning and decision-making algorithms for large-scale machine learning systems, with a particular emphasis on data-efficient fine-tuning of large language and vision-language models under limited or noisy data regimes.

Alongside my academic research, I have experience building and deploying large-scale machine learning systems in industry, including work at Google, Adobe, Airbnb, and SiriusXM. I received my M.S. in Computer Science from UMass Amherst with a GPA of 4.0/4.0. Prior to joining UMass, I earned my B.Sc. in Computer Engineering from Sharif University of Technology, where I was advised by Prof. Ali Sharifi-Zarchi.

CV: Checkout my CV [here].

Email: “a” + [my last name ] at umass dot edu

Research Area

• Online learning - Online decision making
• Trustworthy Machine Learning

Employment

• Machine Learning Data Scientist Intern, Airbnb, Remote, Summer/Fall 2025
  • Developed a framework to detect fake images generated or manipulated by generative models.
  • Designed a strategy to explain the outputs of fake image detection models.
• Machine Learning Data Scientist Intern, SiriusXM and Pandora, Oakland, USA, Summer 2024
  • Developed a framework to explain the similarity of a query and a document retrieved by a search ranking model; Designed on collaboration of multiple LLMs.
  • Enhanced the performance of previous models up to 26% with the final implementation.
• Software Engineer Intern, Google LLC., Mountain View, USA, Spring 2022
  • Developed a highly accurate model to address the interest matching points problem by employing image processing and machine learning techniques, including depth detection, and image perspective detection/modification. Successfully enhanced the performance of SOTA models by up to 50% with the final implementation.
• Machine Learning and Data Science Researcher , Research intern, Adobe, San Jose, USA, Summer 2021
  • Enhanced streaming model of Adobe liquid-mode PDF, ensuring seamless performance in offline/online.
  • Improved quality of experience and wasted bandwidth both up to 20% with the final model.

Publications

• Smoothed Online Optimization for Target Tracking: Robust and Learning-Augmented Algorithms [ALT 2026] [PDF]

  Ali Zeynali, Mahsa Sahebdel, Qingong Liu, Ramesh Sitaraman, Mohammad Hajiesmaili

• Near-Optimal Emission-Aware Online Ride Assignment Algorithm for Peak Demand Hours [Under review] [PDF]

  Mahsa Sahebdel, Ali Zeynali, Noman Bashir, Ramesh Sitaraman, Mohammad Hajiesmaili

• LEAD: Towards Learning-Based Equity-Aware Decarbonization in Ridesharing Platforms [ACM FAccT 2025] [PDF]

  Mahsa Sahebdel, Ali Zeynali, Noman Bashir, Prashant Shenoy, Mohammad Hajiesmaili

• BOLA360: Near-optimal View and Bitrate Adaptation for 360-degree Video Streaming - Journal Version [ACM TOMM] [PDF]

  Ali Zeynali, Mahsa Sahebdel, Mohammad Hajiesmaili, Ramesh Sitaraman

• Robust Learning-Augmented Dictionaries [ICML 2024] [PDF]

  Ali Zeynali, Shahin Kamali, Mohammad Hajiesmaili

• BOLA360: Near-optimal View and Bitrate Adaptation for 360-degree Video Streaming [ACM MMSys 2024] [PDF]

  Ali Zeynali, Mohammad Hajiesmaili, Ramesh Sitaraman

• A Holistic Approach for Equity-aware Carbon Reduction of Ridesharing Platforms [ACM e-Energy 2024] [PDF]

  Mahsa Sahebdel, Ali Zeynali, Noman Bashir, Prashant Shenoy, Mohammad Hajiesmaili

• Data-driven Algorithms for Reducing the Carbon Footprint of Ride-sharing Ecosystems [ACM e-Energy 2023] [PDF]

  Mahsa Sahebdel, Ali Zeynali, Noman Bashir, Mohammad Hajiesmaili, Jimi Oke

• SemEval-2022 Task 8: Multilingual news article similarity [SemEval 2022] [PDF]

  Xi Chen, Ali Zeynali, Chico Camargo, Fabian Flock, Devin Gaffney, Przemyslaw Grabowicz, Scott Hale, David Jurgens, Mattia Samory

• Competitive Algorithms for Online MultidimensionalKnapsack Problems [ACM SIGMETRICS 2022] [PDF]

  Lin Yang, Ali Zeynali, Mohammad Hajiesmaili, Ramesh Sitaraman, Don Towsley

• Data-driven Competitive Algorithms for Online Knapsack and Set Cover [AAAI 2021] [PDF]

  Ali Zeynali, Bo Sun, Mohammad Hajiesmaili, Adam Wierman

• Competitive Algorithms for the Online Multiple KnapsackProblem with Application to Electric Vehicle Charging [ACM SIGMETRICS 2021] [PDF]

  Bo Sun, Ali Zeynali, Tongxin Li, Mohammad Hajiesmaili, Adam Wierman, Danny H.K. Tsang

Selected Projects

• Robust and Unbiased Self-Play Strategy for Fine-Tuning Large Language Models

  A framework for fine-tuning large language models (LLM) on imbalanced datasets

• VSE360: Online 360-degree video streaming simulation evironment [Github]

  Fully simulated python environment to evaluate 360-degree ABR algorithms

• AI-Generated music using Deep Learning + LSTM [Github]

  Generating music using deep learning techniques, and LSTM networks

• AI-Generated short stories using bidirectional LSTM [Github]

  Generating short/tiny stories with deep LSTM

• ZeySed: Deep neural networks for leave classification [Github]

  Classifying image of leaves using deep neural networks

Honors and Awards:

• Thesis Proposal Writing Fellowship Award; Fall 2023 - University of Massachusetts, Amherst

• Nominated for the Microsoft Research Fellowship by the CICS Department; Summer 2021 - University of Massachusetts, Amherst

• Donald F. Towsley Graduate Scholarship; Summer 2021 - University of Massachusetts, Amherst

• Ranked in Top-Ten BSc. students of computer engineering, class of 2019 (among 177 students)

• Rank 24 of 10th IEEEXTREME, 24-hours programming contest among 1823 teams; Fall 2016

• Golden Medalist of 8th International Olympiad IOAA (among more than 200 international students); Summer 2014 - Romania

• Golden Medalist of 9th National Science Olympiad NOAA (among more than 5,000 students); Summer 2013

Contact me:

“a” + [my last name ] at umass dot edu