Ariyo Sanmi, PhD

DATA SCIENTIST | MACHINE LEARNING ENGINEER

ariyo sanmi

About Me

Ariyo Sanmi, PhD

Ariyo is an experienced MLOps Engineer who works with Virgin Media O2. I am responsible for designing, implementing and deploying end-to-end machine learning pipelines in production, including data preprocessing, feature engineering, model training, deployment, monitoring and maintenance.

Ariyo has expertise in containerization and pipeline technologies like Docker, Kubernetes, Kubeflow, Vertex AI and uses them to build scalable and reliable ML infrastructure. He is also proficient in Google Cloud Platform, and has experience in deploying ML models on this platform. 

Ariyo is a team player and is passionate about collaborating with cross-functional teams to solve complex business problems.

Education

PhD: Software Engineering

University of Electronic Science & Technology of China

  • Artificial Intelligence & Deep Learning
2017 - 2020

MSc: Software Engineering

University of Electronic Science & Technology of China

  • Machine Learning and Digital Image Processing
2015 - 2017

B.Eng: Environmental Engineering

Federal University of Technology Akure

  • Design Mechanics and
    Hydrology
2008 - 2013

Experience

MLOps Engineer

Virgin Media O2, UK

  • Design and implement end-to-end machine learning pipelines in production.
  • Manage scalable, reliable, and efficient ML infrastructure on Google Cloud Platform.
  • Collaborate with cross-functional teams - Data Scientists, DevOps Engineers, and other stakeholders to identify business needs.
  • Develop and maintain monitoring and logging systems to track performance and continuously improve the ML systems.
Mar 2022 - Present

ML Engineer

UESTC-IBM Lab, China

  • Supervised and semi-supervised computer vision and pattern recognition.
  • Meteorological analysis for climatic disaster mitigation using deep learning.
  • Design of semantic segmentation and image captioning models with attention neyworks.
July 2020 - Feb 2022

Junior Data Analyst

E-Technograms Technology Services.

  • ETL architectural solutions.
  • Database design and administration.
2014 - 2015

Intern

Cadbury International Limited

  • Engineering analysis and defects detection.
  • Projects monitoring and inspection.
2012 - 2013

Areas of Interest

Data Science

Use of scientific algorithms and processes to extract and analyze data with predictive models.

Data Analysis

Using business intelligence tools to describe insights and patterns in data for business decision-making.

Machine Learning

Automating data analytical modeling using artificial intelligence tools to learn and identify patterns for decision making.

Data Engineering

Building and maintaining data pipeline via practical applications of data collection and validation to ensure a clean, reliable and performative data.

Data Visualization

Providing visual representation of data and findings in an accessible and understandable manner using charts, graphs, maps.

Reinforcement Learning

Applying reinforcement learning techniques to train agents to learn interactive environments via trial and error actions and reward feedbacks..

Computer Vision

The science of understanding and automating the human visual system using high-level artificial intelligence algorithms.

Natural Language Processing

The technology concerned with the understanding and interactions of computers with the human language

Speech Recognition

Automated process of recognizing and the translation of spoken languages and their analyses.

My Hobbies

Sport

Cinema

Traveling

Reading

Guitar

History

FAQ?

MLOps (Machine Learning Operations) is the practice of deploying, managing, and monitoring machine learning models in a production environment. It encompasses a range of activities such as data preparation, model training, model deployment, and ongoing monitoring and maintenance.

Implementing MLOps can lead to several benefits, such as faster time to market for machine learning models, improved model performance and accuracy, better collaboration between data scientists and IT operations teams, and reduced risk of model downtime or errors.

An MLOps infrastructure typically consists of several components, including data storage and management, model training and validation, model deployment and monitoring, and collaboration and communication tools.

An effective MLOps team requires a combination of technical and non-technical skills, such as data engineering, machine learning expertise, software development, and project management. Key roles include data scientists, data engineers, DevOps engineers, and project managers.

Organizations can get started with implementing MLOps by identifying a specific use case or project, establishing a cross-functional team with the necessary skills and expertise, setting up an MLOps infrastructure, and continuously monitoring and improving model performance over time.