Skip to content

Latest commit

 

History

History
108 lines (75 loc) · 4.81 KB

CV_AI.md

File metadata and controls

108 lines (75 loc) · 4.81 KB

Parsa Mohammadi | Curriculum Vitae

Email: [email protected]
Phone: (+98) 9182265891
Personal Website | GitHub | LinkedIn | Google Scholar

EDUCATION

Amirkabir University of Technology (AUT), Tehran, Iran
B.S. in Electrical Engineering | Sep 2020 – Present
GPA: 3.56 (17.42 out of 20) Last 2 years: 3.88 (18.1 out of 20)
Ranked within the top 1% of the Iranian University Entrance exam.

RESEARCH INTERESTS

  • Natural Language Processing (NLP)
  • Large Language Models (LLM)
  • Computer Vision
  • AI for Healthcare

PUBLICATIONS

  • Mohammadi P., Sharifian S. Med Mini-Gemini: Chest X-ray Images Diagnosis and Report Generation. Under preparation.
  • Mohammadi P., Malakouti M., Suratgar A., Menhaj M. Eye Pupil Control Analysis. Vision Research Journal (Under Review). Link

RESEARCH EXPERIENCES

Medical Images Report Generation - HPCRC Lab | Jan 2024

  • Fine-tuned COCA Vision-Language Model achieving 80% accuracy on the MIMIC-CXR dataset.
  • Applied a fine-tuned BERT model achieving an 81% accuracy rate on the MIMIC-III dataset.
  • Supervisor: Dr. Saeed Sharifian (Amirkabir University of Technology, Tehran, Iran)

Tondguyan Petrochemical Company Maintenance System | Jul 2023 – Sep 2023

  • Performed feature reduction on a dataset of mechanical properties of facilities.
  • Implemented the K-means algorithm to identify anomalous data clusters.
  • Supervisor: Dr. Somaye Mohammadi (Sharif University of Technology, Tehran, Iran)

Eye Pupil Control Analysis – DIOR Lab | Oct 2022 – Mar 2023

  • Linearized a complex non-linear model of human eye pupil control to facilitate further analysis.
  • Conducted linear control analysis on the linearized model.
  • Supervisor: Dr. AmirAbolfazl Suratgar (Amirkabir University of Technology, Tehran, Iran)

Data-driven Maintenance of Urban Infrastructure in Smart City | Dec 2021 – Jul 2022

  • Utilized Random Forest, K-means algorithms, and MLP to optimize the cost of maintenance and repair of power networks. Achieved 90% accuracy for predicting power network failures.
  • Developed a Decision Support web application for control centers using the Django framework.

WORK EXPERIENCES

Machine Learning Intern - Asr Gooyesh Pardaz · Full-time, Tehran, Iran | Jul 2023 – Sep 2023

  • Trained an end-to-end Automatic Speech Recognition (ASR) model using E-Branchformer architecture and ESPNet framework on a Persian dataset.
  • Achieved 2.0 WER on Mozilla’s Common Voice dataset (Persian).
  • Implemented the trained model on Hugging Face Space with Gradio, providing an interactive user interface.

TEACHING EXPERIENCES

Amirkabir University of Technology (AUT), Tehran, Iran

  • Logical Circuits | Sep 2024 – Present
    Taught MATLAB and Simulink basics to students.

  • Computer Architecture & Microprocessors | Sep 2023 – Feb 2024
    Created educational materials including projects and weekly homework.

  • Linear Control Systems - Electrical Engineering | Feb 2023 – Jul 2023
    Taught VHDL programming with Vivado and ISE.

  • Sharif University of Technology (SUT), Tehran, Iran
    Automatic Speech Recognition | Sep 2023 – Present
    Conducted weekly quizzes and workshops on ESPNet framework.

SKILLS

  • Programming Skills: Python, TensorFlow, PyTorch, NumPy, Pandas, Sklearn, Gradio, OpenCV, Threading, C/C++, HTML/CSS, JavaScript, Matlab.
  • Machine Learning Skills: NLP, NER, Semantic Analysis, Vision-Language, LLM, NLTK, Speech Processing, ASR, ESPNet, Computer Vision, Segmentation, Object Detection, Classical AI Algorithms.
  • Others: Linux, Git, Docker, Jupyter Notebook, VSCode, LaTeX.

SELECTED PROJECTS

  • Bachelor thesis: Chest X-ray Images Diagnosis and Report Generation | Mar 2024 – Present
    Tuned Mini-Gemini (Pre-trained VLM) on a large dataset of radiology images for VQA task.

  • Radiology Image Diagnosis | Feb 2024 – Mar 2024
    Trained COCA Vision-Language Model on the U-Xray dataset to diagnose TB and pneumonia from chest X-ray images achieving 84.6% accuracy.

  • Cancer detection with chest X-ray CT images | Feb 2024 – Mar 2024
    Fine-tuned YOLOv8 on the Lung-PET-CT-Dx dataset for cancer detection achieving 79% accuracy on the test set.

COURSES

  • Artificial Intelligence [Fall 2021]
  • Machine Learning [Fall 2023]
  • Advanced Programming [Winter 2023]
  • Computer Vision [Winter 2024]

VOLUNTEERING EXPERIENCES

  • Participated in a charity market | Feb 2024
    Created and sold handmade goods to raise funds supporting underprivileged students.

HONORS

  • Won 1st place prize in the IEEE Open Data Hackathon Competition.
  • Ranked among top teams in Irancell Labs AI Hackathon.

LANGUAGE PROFICIENCY

  • English (Exam date: 10/31/2024)
  • Persian (Native)
  • Kurdish (Native)