{
  "$schema": "https://raw.githubusercontent.com/jsonresume/resume-schema/v1.0.0/schema.json",
  "basics": {
    "name": "Amir Mazaheri",
    "label": "Computer Vision Research Scientist · PhD",
    "email": "amirmazaheri1990@gmail.com",
    "phone": "(321) 240-3601",
    "url": "https://amirmazaheri1990.github.io",
    "image": "https://amirmazaheri1990.github.io/profile.jpg",
    "summary": "Computer Vision scientist with deep expertise in large-scale video understanding, Vision-Language Models (VLMs), and multimodal AI systems. Pioneered state-of-the-art solutions across video temporal reasoning, content moderation, in-video search, and multimodal representation learning. PhD from UCF's CRCV. Publications at CVPR, ICCV, ECCV, EMNLP, and AAAI. Multiple US patents.",
    "location": { "city": "Berkeley", "region": "CA", "countryCode": "US" },
    "profiles": [
      { "network": "GitHub", "username": "amirmazaheri1990", "url": "https://github.com/amirmazaheri1990" },
      { "network": "LinkedIn", "username": "amirmazaheri1990", "url": "https://www.linkedin.com/in/amirmazaheri1990/" }
    ]
  },
  "work": [
    {
      "name": "Warner Bros. Discovery (HBO)",
      "location": "San Francisco, CA",
      "position": "Staff Machine Learning Engineer — Computer Vision",
      "startDate": "2025-07",
      "summary": "Leading large-scale video understanding and content moderation systems, including automated harmful content detection and multi-modal safety classifiers for HBO's streaming platform. Developing temporal segmentation, scene/shot boundary detection, VLM-powered video indexing, and LLM-enhanced metadata generation for fine-grained video search and discovery."
    },
    {
      "name": "Tubi",
      "location": "San Francisco, CA",
      "position": "Senior Machine Learning Engineer — Computer Vision",
      "startDate": "2021-09",
      "endDate": "2025-07",
      "summary": "AI-driven content analysis and personalization using VLMs: brand identification, logo detection, scene segmentation, personalized trailers, content artwork generation, and LLM-powered recommendation enhancements."
    },
    {
      "name": "Aibee U.S. Corporation",
      "location": "Palo Alto, CA",
      "position": "Algorithm Scientist",
      "startDate": "2020-07",
      "endDate": "2021-09",
      "summary": "CV models for human tracking in crowds, person re-identification, and AI-based benchmarking tools for tracking."
    },
    {
      "name": "Netflix",
      "location": "Los Gatos, CA",
      "position": "Research Scientist Intern",
      "startDate": "2018-05",
      "endDate": "2018-08",
      "summary": "Semi-supervised activity detection for movies; attentive deep neural network for movie billboard design optimization."
    },
    {
      "name": "Nielsen",
      "location": "Orlando, FL",
      "position": "Research Fellowship",
      "startDate": "2017-06",
      "endDate": "2018-08",
      "summary": "Automated TV advertisement analysis: classification, tagging, and descriptive analysis of broadcast ads."
    },
    {
      "name": "Center for Research in Computer Vision (CRCV), University of Central Florida",
      "location": "Orlando, FL",
      "position": "Graduate Research Assistant",
      "startDate": "2013-08",
      "endDate": "2020-04",
      "summary": "PhD research with Prof. Mubarak Shah on video understanding, visual question answering, Visual Text Correction, video generation, and multimodal AI."
    }
  ],
  "education": [
    {
      "institution": "University of Central Florida",
      "area": "Computer Science",
      "studyType": "Ph.D.",
      "endDate": "2020-04",
      "score": "Dissertation: Video Content Understanding Using Text. Advisor: Prof. Mubarak Shah, CRCV."
    },
    {
      "institution": "University of Central Florida",
      "area": "Computer Science",
      "studyType": "M.Sc.",
      "endDate": "2016-05"
    },
    {
      "institution": "Sharif University of Technology",
      "area": "Computer Science",
      "studyType": "B.S.",
      "endDate": "2013-07"
    }
  ],
  "publications": [
    { "name": "WoundNet: A Domain-Adaptable Few-Shot Classification Framework for Wound Healing Assessment", "publisher": "ISBI", "releaseDate": "2023" },
    { "name": "Context-Aware Analysis of Group Submissions for Group Anomaly Detection and Performance Prediction", "publisher": "AAAI", "releaseDate": "2023" },
    { "name": "Video Generation from Text Employing Latent Path Construction for Temporal Modeling", "publisher": "ICPR", "releaseDate": "2022" },
    { "name": "MMFT-BERT: Multimodal Fusion Transformer with BERT Encodings for Visual Question Answering", "publisher": "Findings of EMNLP", "releaseDate": "2020" },
    { "name": "Deep Photo Cropper and Enhancement", "publisher": "ICIP", "releaseDate": "2020" },
    { "name": "Pay Attention! — Robustifying a Deep Visuomotor Policy through Task-Focused Attention", "publisher": "CVPR", "releaseDate": "2019" },
    { "name": "Visual Text Correction", "publisher": "ECCV", "releaseDate": "2018", "url": "https://amirmazaheri1990.github.io/VTC/" },
    { "name": "Learning a Multi-concept Video Retrieval Model with Multiple Latent Variables", "publisher": "ACM TOMM", "releaseDate": "2018" },
    { "name": "Video Fill In the Blank using LR/RL LSTMs with Spatial-Temporal Attentions", "publisher": "ICCV", "releaseDate": "2017" }
  ],
  "skills": [
    { "name": "Programming Languages", "keywords": ["Python", "C/C++", "SQL", "MATLAB", "Scala", "Java"] },
    { "name": "ML Frameworks", "keywords": ["PyTorch", "TensorFlow", "Keras"] },
    { "name": "VLMs & LLMs", "keywords": ["Vision-Language Models", "LLMs", "instruction tuning", "multimodal fine-tuning"] },
    { "name": "Cloud & Big Data", "keywords": ["AWS", "Apache Spark", "PySpark"] },
    { "name": "DevOps", "keywords": ["Docker", "CI/CD"] }
  ],
  "meta": {
    "canonical": "https://amirmazaheri1990.github.io/cv.json",
    "version": "2026",
    "lastModified": "2026-04-18"
  }
}