Yhary Estefania Arias Trillos
Download
Recommend
Messages
Share
Yhary Estefania Arias Trillos

Yhary Estefania Arias Trillos

About

Detail

Machine Learning Engineering
Colombia
Contact Yhary regarding: 
work
Full-time jobs
Starting at USD2.5K/month
groups
Networking

Timeline

work
Job
school
Education
folder
Project (professional or personal)
flag
Award
auto_stories
Publication

Résumé

Jobs verified_user 0% verified
  • P
    Data Science Teacher
    Pontificia Universidad Javeriana Cali
    Aug 2023 - Sep 2023 (2 months)
    Python programming teacher applied to data science.
  • A
    MLOps
    Applaudo
    Dec 2022 - Apr 2023 (5 months)
    Skills: Cloud Computing · Google Cloud Platform (GCP) · Google BigQuery · Git
  • Applaudo Studios
    Machine Learning Engineering
    Applaudo Studios
    May 2022 - Apr 2023 (1 year)
    I was part of the seven-person team that designed a CNN image recognition model to be used by the Miami Heat basketball team to assess its game strategies. We analyzed 137 videos (3 years) extracting frames per second and trained the models to identify the player’s actions considering the distances between the ball and the basket/player. I designed and developed a pilot for a ChatBot using BERT to apply LLM techniques. The data was explored and pre-processed to feed the BERT model where I applied the fine- tuning technique. The pilot showed the potential to improve 70% the quality of responses. I lead proof of concept (internal project) to assess emotion detection by assessing the eye’s iris of job interview candidates. We selected 200 inte
  • S
    Contributor
    Somos NLP
    Mar 2022 - Current (3 years 5 months)
    Active contributor to the Somos NLP community https://somosnlp.org/comunidad
  • U
    Freelance - Computer Vision
    Updata
    Feb 2022 - Mar 2022 (2 months)
    Development of a computer vision algorithm that identified moving objects in order to obtain an estimated weight of the same taking into account certain characteristics that were labeled manually due to the complexity with a video labeling tool. Docker containers were used to automate certain processes. Used SnowFlake as a key platform to manage and analyze machine learning projects with large data sets. This integration allowed for efficient storage and processing, facilitating data-driven decision making and improving the effectiveness of the implemented models.
  • Universidad Industrial de Santander
    UIS Teacher - MisionTIC 2022 Project
    Universidad Industrial de Santander
    Sep 2021 - Dec 2022 (1 year 4 months)
    I participated in the orientation and training as a teacher of the object-oriented programming course with Python and Java, software development and its requirements. I made use of the Spring Boot framework to create projects with microservices, use of APIs, connection to the database and the university server to deploy the applications developed by each student. I developed a Git course focused on the use of terminal commands for collaborative work as an important basis for project development. For this we used GitLab. In the mobile development cycle, we made use of the Java programming language for Android Studio, as for the connection to the database, we used FireBase, developed projects with geolocation, use of intents, menus, dialogs
  • Sistemas y Computadores SA
    Innovation and components engineer
    Sistemas y Computadores SA
    Jan 2020 - Oct 2020 (10 months)
    Development of Machine Learning algorithms for data processing in identification documents and forms. I worked developing with Python, managing queries and database connection with SQL, using docker to automate the deployment of the developed models. I developed a chatbot applying the BETO model for NLP that contains a large corpus in Spanish, to automate and provide an effective service to each customer.
  • S
    Junior developer of mobile (intern)
    Switch 2 Actions
    Jun 2019 - Dec 2019 (7 months)
    Junior developer of mobile applications with Flutter, database connection with SQL and UI and functionality testing.
  • Escuela Colombiana de Inteligencia Artificial
    Founder
    Escuela Colombiana de Inteligencia Artificial
    Feb 2019 - Current (6 years 6 months)
    I teach once a week on topics of basic programming with Python, artificial intelligence models, data preprocessing and NLP model development. I do advertising in social networks.
Education verified_user 0% verified
  • Universitat Pompeu Fabra
    PhD in Information and Communication Technologies - AI, PhD in Information and Communication Technologies, focused on Ar
    Universitat Pompeu Fabra
    Oct 2023 - Current (1 year 10 months)
    Skills: Artificial Intelligence (AI)
  • Universitat Pompeu Fabra
    PhD in Information and Communication Technologies - AI, PhD in Information and Communication Technologies, focused on Ar
    Universitat Pompeu Fabra
    Oct 2023 - Current (1 year 10 months)
    Skills: Artificial Intelligence (AI)
  • Universitat Pompeu Fabra
    PhD in Information and Communication Technologies - AI, PhD in Information and Communication Technologies, focused on Ar
    Universitat Pompeu Fabra
    Oct 2023 - Current (1 year 10 months)
    Skills: Artificial Intelligence (AI)
  • U
    Master in artificial intelligence, shape recognition and digital image., computer engineering
    Universitat Politcnica de Valncia UPV
    Sep 2020 - Apr 2022 (1 year 8 months)
    Carolina Foundation Fellow, My master's thesis project consists of a computer vision model for classification of thyroid ultrasound images into 5 classes based on the EU-TIRADS screening test.
  • F
    VIII cycle of special courses, faculty of mathematics, NLP for decision making in organizations and e-commerce analytics
    Fundacin Universitaria Konrad Lorenz
    Nov 2019
    Natural language processing for decision making in organizations and e-commerce analytics with R and python
  • F
    VIII cycle of special courses, faculty of mathematics, NLP for decision making in organizations and e-commerce analytics
    Fundacin Universitaria Konrad Lorenz
    Nov 2019
    Natural language processing for decision making in organizations and e-commerce analytics with R and python
  • F
    VIII cycle of special courses, faculty of mathematics, NLP for decision making in organizations and e-commerce analytics
    Fundacin Universitaria Konrad Lorenz
    Nov 2019
    Natural language processing for decision making in organizations and e-commerce analytics with R and python
  • Universidad Industrial de Santander
    Certificado de participación como ponente en el evento "IDEAS PARA TRANSFORMAR EL MUNDO, UIS Fest, Investigación de oper
    Universidad Industrial de Santander
    Jan 2019 - Current (6 years 7 months)
    Por haber participado como ponente con el poster titulado: “Parametrización de un modelo de red neuronal convolucional para la clasificación de cáncer en ecografías de tiroides”, en el evento U18 FEST – IDEAS PARA TRANSFORMAR EL MUNDO, realizado en la ciudad de Bucaramanga del 23 al 28 de septiembre de 2019.
  • A
    Tercer puesto. Premio a protocolos de investigación, Medicina, Ingeniería
    Asociacin Colombiana de Radiologa
    Jan 2019 - Dec 2019 (1 year)
    La Asociación Colombiana de Radiología hace constar que el grupo investigador: Boris Marconi. Federico Lubinus. Yhary Arias. Ariel Ortiz. Cesar Rueda. Universidad Autónoma de Bucaramanga - Foscal. Obtuvo reconocimiento académico en el 44° Congreso Colombiano de Radiología y 13° Congreso de la Sociedad Iberoamericana de Imagen Mamaria, realizados del 8 al 10 de agosto de 2019 en Cartagena – Colombia. Por el Protocolo de investigación titulado: Protocolo de implementación de un sistema de Inteligencia Artificial (IA) basado en Redes Neuronales Convolucionadas (CNN) como asistencia diagnóstica para la detección y clasificación de nódulo tiroideo por ultrasonido Tercer puesto. Premio a Protocolos de Investigación
  • Universidad Industrial de Santander
    Certificado de participación como ponente en el evento "IDEAS PARA TRANSFORMAR EL MUNDO, UIS Fest, Investigación de oper
    Universidad Industrial de Santander
    Jan 2019 - Current (6 years 7 months)
    Por haber participado como ponente con el poster titulado: “Parametrización de un modelo de red neuronal convolucional para la clasificación de cáncer en ecografías de tiroides”, en el evento U18 FEST – IDEAS PARA TRANSFORMAR EL MUNDO, realizado en la ciudad de Bucaramanga del 23 al 28 de septiembre de 2019.
  • G
    Programa de aceleración en Inteligencia Artificial, Inteligencia artificial
    Google pour les pros
    Jan 2019 - Current (6 years 7 months)
    Ruta N: Programa de aceleración en Inteligencia Artificial Apoya: Google Cloud Con este proyecto de investigación se busca mejorar la capacidad con la cual los doctores encargados de los accidentes cerebrovasculares, pueden realizar diagnósticos sobre las lesiones que el paciente sufre, esto con el fin de mejorar los tiempos de respuesta a una lesión, que en muchos casos puede ser vital para salvar la vida del paciente.
  • Universidad Industrial de Santander
    Certificado de participación como ponente en el evento "IDEAS PARA TRANSFORMAR EL MUNDO, UIS Fest, Investigación de oper
    Universidad Industrial de Santander
    Jan 2019 - Current (6 years 7 months)
    Por haber participado como ponente con el poster titulado: “Parametrización de un modelo de red neuronal convolucional para la clasificación de cáncer en ecografías de tiroides”, en el evento U18 FEST – IDEAS PARA TRANSFORMAR EL MUNDO, realizado en la ciudad de Bucaramanga del 23 al 28 de septiembre de 2019.
  • A
    Panelista, Medicina
    Asociacin Colombiana de Radiologa
    Jan 2019 - Dec 2019 (1 year)
    La Asociación Colombiana de Radiología hace constar que: Yhary Estefanía Arias Participó como panelista invitada en el: Curso de Inteligencia Artificial “Aplicación en el Diagnóstico por Imagen” Realizado en Medellín - Colombia, los días 15 y 16 de noviembre de 2019
  • U
    Ingeniera de Sistemas, Ingeniería de software
    Universidad Autnoma de Bucaramanga
    Jan 2016 - Dec 2019 (4 years)
    Degree project (Meritorious) Project name: "Web system for recognition and classification of pathologies through medical images based on machine learning techniques" Active participation in the training process for research in the MEDIALAB seedbed (Certificate) - Segmentation of lesions in cerebrovascular accidents (Participation in the artificial intelligence acceleration program, developed by RUTA N, C4IR.CO and GOOGLE).
  • Universidad Francisco de Paula Santander
    Certificación como ponente, Investigación de operaciones/operativa
    Universidad Francisco de Paula Santander
    Congreso internacional de innovación y aprobación de las tecnologías de la información y las comunicaciones, CIINATIC 2019
Projects (professional or personal) verified_user 0% verified
  • Sistemas y Computadores SA
    Exploration/query of large datasets for chatbot development.
    Sistemas y Computadores SA
    Jan 2020 - Oct 2023 (3 years 10 months)
Awards verified_user 0% verified
  • U
    Master's scholarship
    Universidad Politcnica de Valencia
    Apr 2022
Publications verified_user 0% verified
  • C
    Redes neuronales convolucionales: un modelo de Deep Learning en imágenes diagnósticas.
    Colombian Journal of Radiology ACR
    Sep 2021
    Convolutional neural networks, a model for deep learning in diagnostic imaging. A topic review. Advances in artificial intelligence have impacted several areas of everyday life, as well as in the area of medicine. Due to the rapid application of deep learning in biomedical data, radiological and nuclear imaging has begun to adopt this technique. Deep learning is expected to have an effect on the process of image acquisition and interpretation, as well as on decision making. This review first provides an overview of the basic concepts and operation of convolutional neural networks, as well as current insights into the medical application focused on diagnostic imaging.