About Me
Welcome to my website. My name is Ali Bidaran, and I am an AI engineer. I studied electronic engineering, while my main profession is designing ML and AI models for various projects. My main research interest is designing Computer Vision and NLP, in particular LLMs, for different applications. Moreover, I often contribute designing embedded systems, PCB boards and IoT projects.
Skills
AI and Machine learning
- Deep learning (frameworks like PyTorch, ONNX and Keras)
- Machine Learning (frameworks like Scikit-learn)
- Computer Vision (frameworks such as OpenCV, Diffusers and OpenVino)
- NLP and LLMs (frameworks Transformers, Gensim, Spacy)
Front-End
- HTML5
- CSS3
- JavaScript
Back-End
- Programming Languages (e.g., Python)
- Web Frameworks (e.g.,Flask)
- Databases (e.g., MySQL, MongoDB)
- Fast prototyping (e.g., QT and Gradio)
DevOps
- Version Control (Git)
- Deployment Tools (e.g., Docker, AWS)
- CI/CD Pipelines
Data Processing
- Programming Languages (e.g., Python)
- Pandas
- Numpy
- Data visualization (e.g. Plotly, Matplotlib and Dash)
Embedded Systems and IoT
- Programming Languages (e.g.,C/C++, Micropython)
- Platforms (e.g. Arduino IDE, Mbed)
- Boards (e.g, Arduino, STM32, ESP32, Raspberry pi)
- PCB board designing (e.g. Kicad)
Education
Amirkabir University of Technology (Tehran, Iran)
Master's of Electrical Engineering-Digital Electronic Systems
Islamic Azad University (Central Tehran Branch)
Bachelor degree in Electrical Engineering
Summer Schools and Conferences
Oxford University, England
Attendance of Oxford Machine Learning School (OXML)(2024)
Ukrainian Catholic University, Ukraine
Lviv Data Science Summer School (2022)
12th International Conference on Information and Knowledge Technology, Iran
Contribution as Invited speaker (2021)
Gdansk University of Technology, Poland
Attending at International summer school on deep learning (ISSonDL)(2020)
Politehnica University of Bucharest, Romania
Attending at International Summer School of Imaging and Medical Application (SSIMA) (2019)
Publications
Projects
Instruction2PythonCode
![Project 1](https://huggingface.co/datasets/huggingface/brand-assets/resolve/main/hf-logo.png)
Fine-tuning the latest Google’s open source LLM model (Gemma) to generate Python codes for various applications, from Machine learning to backend and web development.
View ProjectVirtual Doctor
![Project 1](https://huggingface.co/datasets/huggingface/brand-assets/resolve/main/hf-logo.png)
Implementing and fine-tuning the latest LLM model, in particular Gemma, being a member of Google’s Gemini AI, for medical and healthcare applications.
View ProjectDavinci Eye
Designing an Unet model for segmenting surgery images and optimizing its weights to get real-time performance on the Intel CPU families.Designing a web app with the trained model with Python, Numpy, OpenVino, ONNX, and Gradio. Deploying it in HuggingFace spaces(AWS)
Medical transcript generator
An open source project that contains the implementation of the medical transcript generator by fine-tuning a LLM such as GPT-2 to create realistic medical reports with Python, PyTorch, and Transformers libraries. Having been downloaded over 18 thousands since March 2023.
Symptom2Disease
Fine-tuning a LLM such as BERT to detect 24 different diseases from textual symptoms in Python, PyTorch, and Transformers.
Alzaimer MRI detection
Designing a CNN model and optimizing it for CPUs to different stages of Alzaimer from MRI images.
General Image Captioning
Implementing an image captioning model based on GPT and CNN models to create realistic captions for images with Python, PyTorch, and Transformers.
Contact
Feel free to reach out to me from these platforms: