Personal Website

Giannis Kousis

AI Engineer

Contact me

About me

I am an experienced AI Engineer with a strong background in developing and deploying scalable AI solutions. My adaptability and strong work ethic have been key to my success. I excel at delivering rapid solutions for urgent projects and can also focus on providing high-quality, detailed implementations when required, ensuring I meet diverse project needs effectively.


Passionate about the limitless potential of AI, I'm always open to connecting with professionals and exploring opportunities in this domain.

Work experience

Founder
Staysious, Greece
01/2024 - Present

  • Developed AI-driven solutions to minimize repetitive queries and improve overall guest satisfaction for Airbnb hosts.
  • Acquired comprehensive expertise in project planning, operational efficiency, and strategic visioning.
  • Designed and deployed a mobile app using Google Firebase for backend services and Flutter for front-end development.

    • Team Lead
    IridaLabs, Greece
    10/2022 - 12/2023

  • Led the development of PerCV.ai, end-to-end Vision AI platform, starting with a small but high-skilled team of three engineers and collaborating extensively with multiple internal teams to achieve project goals.
  • Designed comprehensive cloud infrastructure architecture using a mix of AWS services, spot instances, and on-premises computing resources, enhancing system efficiency and scalability while significantly reducing operational costs.
  • Facilitated daily communication and collaboration, efficiently addressing engineering challenges in a dynamic startup environment.

    • Machine Learning Engineer
    IridaLabs, Greece
    08/2019 - 10/2022

  • Managed the complete AI/ML pipeline for top clients, covering all stages of ML projects from data handling and model development to deployment and monitoring.
  • Demonstrated strong critical thinking and problem-solving skills to ensure efficient model performance in real-world applications.

    • Education

    Integrated Master's in Computer Engineering (B.Sc. + M.Sc.)
    Greece , University of Patras
    2022

    POS4work Restart-Up , Entrepreneurship School of Patras
    Greece
    2018

    Papers

    Deep Learning Methods for Accurate Skin Cancer Recognition and Mobile Application
    M.Sc. Thesis, University of Patras

    (Kousis, I., Perikos, I., Hatzilygeroudis, I., & Virvou, M.) Electronics, 11(9), 1294.
    https://www.mdpi.com/2079-9292/11/9/1294

    This paper presents a comprehensive study of 11 CNN architectures for skin cancer detection using the HAM10000 dataset. The best model, DenseNet169, achieved 92.25% accuracy and was implemented in a mobile app for real-time classification of skin lesions. The app also provides UV exposure guidance based on user skin type and sunscreen used.

    Skills

    Discover the proficiencies that allow me to create Advanced Machine Learning and AI Solutions.

    Machine Learning Model Development

    Expertise in creating and refining machine learning algorithms.

    Data Collection and Annotation

    Proficiency in gathering and labeling data for AI models.

    AI Model Optimization

    Skilled in enhancing the performance of AI models.

    AI Deployment

    Experience in deploying AI solutions in cloud and edge environments.

    AWS Services

    Knowledgeable in utilizing Amazon Web Services for cloud computing and storage needs.

    Programming Languages

    Proficient in various programming languages essential for software and AI development.

    Personal Projects

    Project 1:
    Geodeals App (2017)

    Description:

    A project to find stores and offers nearby. It consisted of three different applications (customer, store, admin). (Android, Firebase, Android Studio, Google Maps API)

    Project 2:
    (IoT) Agriculture Monitor Station (2021)

    Description:

    ESP32 based microcontroler that is able to collect environmental conditions like temperature and humidity and sends the collected data through a GSM network on a cloud dashboard hosted on AWS EC2 instance. The system is powered with renewable energy from a solar panel and a rechargeable battery. (Arduino, AWS )

    Project 3:
    SkinHealth App (BSc Thesis) (2019)

    Description:

    An application developed for early skin cancer diagnosis through images from smartphones devices using Deep Learning. (TensorBoard, TensorFlow Lite, Keras, Android, Python, Firebase) https://www.mdpi.com/2079-9292/11/9/1294