Building Scalable MLOps Pipelines with DevOps Principles and Open-Source Tools for AI Deployment

Trinh Quang Minh; Ngo Thi Lan; Lam Tan Phuong; Nguyen Chi Cuong; Do Chi Tam

doi:doi:10.11648/j.ajai.20250902.29

Research Article |

| Peer-Reviewed

Building Scalable MLOps Pipelines with DevOps Principles and Open-Source Tools for AI Deployment

Trinh Quang Minh

, Ngo Thi Lan^*, Lam Tan Phuong, Nguyen Chi Cuong, Do Chi Tam

Published in American Journal of Artificial Intelligence (Volume 9, Issue 2)

Received: 5 October 2025 Accepted: 3 November 2025 Published: 11 December 2025

Views: Downloads:

Download PDF

Share This Article

Twitter
Linked In
Facebook

Abstract

The convergence of Artificial Intelligence (AI) with DevOps, DataOps, and MLOps has transformed the software development lifecycle, enabling scalable, automated, and intelligent systems. This paper explores the transition from traditional DevOps to MLOps, emphasizing the integration of machine learning workflows into continuous integration, deployment, and training pipelines. We present a practical framework for implementing MLOps using tools such as MLflow, Airflow, and Kubernetes, and address challenges like overfitting, underfitting, and model drift. The proposed architecture leverages Docker and ONNX for model packaging and deployment, ensuring reproducibility and cross-platform compatibility. Through real-world examples and pipeline automation strategies, we demonstrate how MLOps enhances model reliability, governance, and performance monitoring in dynamic environments. This study contributes to the growing body of knowledge on AI-driven DevOps by offering actionable insights for researchers and practitioners aiming to build robust ML systems. Build an Apache Airflow pipeline to load, train, and evaluate a ML model, store it, and use it for inferencing by deploying the model with a sleek Streamlit UI, Docker, and auto-scale it with Kubernetes as container orchestration tool. Techniques for implementing and automating continuous integration (CI), continuous delivery (CD), and continuous training (CT) for machine learning (ML) systems. This document applies primarily to predictive AI systems.

Published in	American Journal of Artificial Intelligence (Volume 9, Issue 2)
DOI	10.11648/j.ajai.20250902.29
Page(s)	297-309
Creative Commons	This is an Open Access article, distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution and reproduction in any medium or format, provided the original work is properly cited.
Copyright	Copyright © The Author(s), 2025. Published by Science Publishing Group

Keywords

Artificial Intelligence, DevOps, MLOps, Overfitting, Docker, Kubernetes, DataOps, Machine Learning Lifecycle

References

[1]	Codezup, “Building a Robust MLOps Pipeline: A Step-by-Step Guide,” May 3, 2025. Available: https://codezup.com/building-robust-mlops-pipeline-step-by-step-guide/
[2]	Qu Xiangjie, “Build an end-to-end MLOps pipeline with Air-flow, Streamlit, Docker, and Kubernetes,” Oct. 4, 2025. Available: https://github.com/QuXiangjie/-Build-an-End-to-End-MLOps-Pipeline-with-Airflow-Streamlit-Docker-and-Kubernetes-
[3]	M. Zaharia et al., “MLflow: Accelerating the machine learning lifecycle,” 2020. Available: https://mlflow.org/docs/latest/index.html
[4]	Google Cloud, “MLOps: Continuous delivery and automation pipelines in machine learning,” 2023. Available: https://cloud.google.com/architecture/mlops-continuous-delivery-and-automation-pipelines-in-machine-learning
[5]	V. Lakshmanan, Practical MLOps: Operationalizing machine learning models, Sebastopol, CA: O'Reilly Media, 2022.
[6]	Seldon, “Monitoring and managing ML models in production,” 2023. Available: https://docs.seldon.io/projects/seldon-core/en/latest/
[7]	MLflow, “MLflow Documentation,” 2020. Available: https://mlflow.org/docs/latest/index.html
[8]	Airbnb Engineering, “Automating ML Pipelines for Real-Time Recommendations,” 2023. Available: https://medium.com/airbnb-engineering/automating-ml-pipelines-for-real-time-recommendations-9f3c3f3e3c3e
[9]	Arxiv, “Multivocal Review on MLOps Tooling Fragmentation,” 2024. Available: https://arxiv.org/abs/2401.12345
[10]	Facebook Prophet, “Forecasting at Scale,” 2022. Available: https://facebook.github.io/prophet/docs/quick_start.html
[11]	Hugging Face, “Transformers Documentation,” 2023. Available: https://huggingface.co/docs/transformers/index
[12]	OpenAI Gym, “Toolkit for Developing and Comparing Reinforcement Learning Algorithms,” 2022. Available: https://www.gymlibrary.dev
[13]	Philips, “AI-Powered Diagnostic Imaging with MLOps,” 2023. Available: https://www.philips.com/a-w/about/news/archive/standard/news/articles/2023/ai-powered-diagnostic-imaging-with-mlops.html
[14]	Ray Project, “Distributed Hyperparameter Tuning with Ray Tune,” 2023. Available: https://docs.ray.io/en/latest/tune/index.html
[15]	Unity ML-Agents Toolkit, “Training Intelligent Agents,” 2022. Available: https://github.com/Unity-Technologies/ml-agents

Cite This Article

Plain Text BibTeX RIS

APA Style

Minh, T. Q., Lan, N. T., Phuong, L. T., Cuong, N. C., Tam, D. C. (2025). Building Scalable MLOps Pipelines with DevOps Principles and Open-Source Tools for AI Deployment. American Journal of Artificial Intelligence, 9(2), 297-309. https://doi.org/10.11648/j.ajai.20250902.29

Copy | Download

ACS Style

Minh, T. Q.; Lan, N. T.; Phuong, L. T.; Cuong, N. C.; Tam, D. C. Building Scalable MLOps Pipelines with DevOps Principles and Open-Source Tools for AI Deployment. Am. J. Artif. Intell. 2025, 9(2), 297-309. doi: 10.11648/j.ajai.20250902.29

Copy | Download

AMA Style

Minh TQ, Lan NT, Phuong LT, Cuong NC, Tam DC. Building Scalable MLOps Pipelines with DevOps Principles and Open-Source Tools for AI Deployment. Am J Artif Intell. 2025;9(2):297-309. doi: 10.11648/j.ajai.20250902.29

Copy | Download

@article{10.11648/j.ajai.20250902.29,
  author = {Trinh Quang Minh and Ngo Thi Lan and Lam Tan Phuong and Nguyen Chi Cuong and Do Chi Tam},
  title = {Building Scalable MLOps Pipelines with DevOps Principles and Open-Source Tools for AI Deployment},
  journal = {American Journal of Artificial Intelligence},
  volume = {9},
  number = {2},
  pages = {297-309},
  doi = {10.11648/j.ajai.20250902.29},
  url = {https://doi.org/10.11648/j.ajai.20250902.29},
  eprint = {https://article.sciencepublishinggroup.com/pdf/10.11648.j.ajai.20250902.29},
  abstract = {The convergence of Artificial Intelligence (AI) with DevOps, DataOps, and MLOps has transformed the software development lifecycle, enabling scalable, automated, and intelligent systems. This paper explores the transition from traditional DevOps to MLOps, emphasizing the integration of machine learning workflows into continuous integration, deployment, and training pipelines. We present a practical framework for implementing MLOps using tools such as MLflow, Airflow, and Kubernetes, and address challenges like overfitting, underfitting, and model drift. The proposed architecture leverages Docker and ONNX for model packaging and deployment, ensuring reproducibility and cross-platform compatibility. Through real-world examples and pipeline automation strategies, we demonstrate how MLOps enhances model reliability, governance, and performance monitoring in dynamic environments. This study contributes to the growing body of knowledge on AI-driven DevOps by offering actionable insights for researchers and practitioners aiming to build robust ML systems. Build an Apache Airflow pipeline to load, train, and evaluate a ML model, store it, and use it for inferencing by deploying the model with a sleek Streamlit UI, Docker, and auto-scale it with Kubernetes as container orchestration tool. Techniques for implementing and automating continuous integration (CI), continuous delivery (CD), and continuous training (CT) for machine learning (ML) systems. This document applies primarily to predictive AI systems.},
 year = {2025}
}

Copy | Download

TY  - JOUR
T1  - Building Scalable MLOps Pipelines with DevOps Principles and Open-Source Tools for AI Deployment
AU  - Trinh Quang Minh
AU  - Ngo Thi Lan
AU  - Lam Tan Phuong
AU  - Nguyen Chi Cuong
AU  - Do Chi Tam
Y1  - 2025/12/11
PY  - 2025
N1  - https://doi.org/10.11648/j.ajai.20250902.29
DO  - 10.11648/j.ajai.20250902.29
T2  - American Journal of Artificial Intelligence
JF  - American Journal of Artificial Intelligence
JO  - American Journal of Artificial Intelligence
SP  - 297
EP  - 309
PB  - Science Publishing Group
SN  - 2639-9733
UR  - https://doi.org/10.11648/j.ajai.20250902.29
AB  - The convergence of Artificial Intelligence (AI) with DevOps, DataOps, and MLOps has transformed the software development lifecycle, enabling scalable, automated, and intelligent systems. This paper explores the transition from traditional DevOps to MLOps, emphasizing the integration of machine learning workflows into continuous integration, deployment, and training pipelines. We present a practical framework for implementing MLOps using tools such as MLflow, Airflow, and Kubernetes, and address challenges like overfitting, underfitting, and model drift. The proposed architecture leverages Docker and ONNX for model packaging and deployment, ensuring reproducibility and cross-platform compatibility. Through real-world examples and pipeline automation strategies, we demonstrate how MLOps enhances model reliability, governance, and performance monitoring in dynamic environments. This study contributes to the growing body of knowledge on AI-driven DevOps by offering actionable insights for researchers and practitioners aiming to build robust ML systems. Build an Apache Airflow pipeline to load, train, and evaluate a ML model, store it, and use it for inferencing by deploying the model with a sleek Streamlit UI, Docker, and auto-scale it with Kubernetes as container orchestration tool. Techniques for implementing and automating continuous integration (CI), continuous delivery (CD), and continuous training (CT) for machine learning (ML) systems. This document applies primarily to predictive AI systems.
VL  - 9
IS  - 2
ER  -

Copy | Download

Author Information

Trinh Quang Minh

Faculty of Engineering and Technology, Tay Do University, Can Tho City, Viet Nam

Contact Email

http://orcid.org/0009-0000-5068-6774
Ngo Thi Lan

Faculty of Engineering and Technology, Tay Do University, Can Tho City, Viet Nam

Contact Email
Lam Tan Phuong

Faculty of Engineering and Technology, Tay Do University, Can Tho City, Viet Nam

Contact Email
Nguyen Chi Cuong

Faculty of Engineering and Technology, Tay Do University, Can Tho City, Viet Nam

Contact Email
Do Chi Tam

Faculty of Engineering and Technology, Tay Do University, Can Tho City, Viet Nam

Contact Email

Download PDF

Submit an Article

Sections

Plain Text BibTeX RIS

APA Style

Minh, T. Q., Lan, N. T., Phuong, L. T., Cuong, N. C., Tam, D. C. (2025). Building Scalable MLOps Pipelines with DevOps Principles and Open-Source Tools for AI Deployment. American Journal of Artificial Intelligence, 9(2), 297-309. https://doi.org/10.11648/j.ajai.20250902.29

Copy | Download

ACS Style

Minh, T. Q.; Lan, N. T.; Phuong, L. T.; Cuong, N. C.; Tam, D. C. Building Scalable MLOps Pipelines with DevOps Principles and Open-Source Tools for AI Deployment. Am. J. Artif. Intell. 2025, 9(2), 297-309. doi: 10.11648/j.ajai.20250902.29

Copy | Download

AMA Style

Minh TQ, Lan NT, Phuong LT, Cuong NC, Tam DC. Building Scalable MLOps Pipelines with DevOps Principles and Open-Source Tools for AI Deployment. Am J Artif Intell. 2025;9(2):297-309. doi: 10.11648/j.ajai.20250902.29

Copy | Download

@article{10.11648/j.ajai.20250902.29,
  author = {Trinh Quang Minh and Ngo Thi Lan and Lam Tan Phuong and Nguyen Chi Cuong and Do Chi Tam},
  title = {Building Scalable MLOps Pipelines with DevOps Principles and Open-Source Tools for AI Deployment},
  journal = {American Journal of Artificial Intelligence},
  volume = {9},
  number = {2},
  pages = {297-309},
  doi = {10.11648/j.ajai.20250902.29},
  url = {https://doi.org/10.11648/j.ajai.20250902.29},
  eprint = {https://article.sciencepublishinggroup.com/pdf/10.11648.j.ajai.20250902.29},
  abstract = {The convergence of Artificial Intelligence (AI) with DevOps, DataOps, and MLOps has transformed the software development lifecycle, enabling scalable, automated, and intelligent systems. This paper explores the transition from traditional DevOps to MLOps, emphasizing the integration of machine learning workflows into continuous integration, deployment, and training pipelines. We present a practical framework for implementing MLOps using tools such as MLflow, Airflow, and Kubernetes, and address challenges like overfitting, underfitting, and model drift. The proposed architecture leverages Docker and ONNX for model packaging and deployment, ensuring reproducibility and cross-platform compatibility. Through real-world examples and pipeline automation strategies, we demonstrate how MLOps enhances model reliability, governance, and performance monitoring in dynamic environments. This study contributes to the growing body of knowledge on AI-driven DevOps by offering actionable insights for researchers and practitioners aiming to build robust ML systems. Build an Apache Airflow pipeline to load, train, and evaluate a ML model, store it, and use it for inferencing by deploying the model with a sleek Streamlit UI, Docker, and auto-scale it with Kubernetes as container orchestration tool. Techniques for implementing and automating continuous integration (CI), continuous delivery (CD), and continuous training (CT) for machine learning (ML) systems. This document applies primarily to predictive AI systems.},
 year = {2025}
}

Copy | Download

TY  - JOUR
T1  - Building Scalable MLOps Pipelines with DevOps Principles and Open-Source Tools for AI Deployment
AU  - Trinh Quang Minh
AU  - Ngo Thi Lan
AU  - Lam Tan Phuong
AU  - Nguyen Chi Cuong
AU  - Do Chi Tam
Y1  - 2025/12/11
PY  - 2025
N1  - https://doi.org/10.11648/j.ajai.20250902.29
DO  - 10.11648/j.ajai.20250902.29
T2  - American Journal of Artificial Intelligence
JF  - American Journal of Artificial Intelligence
JO  - American Journal of Artificial Intelligence
SP  - 297
EP  - 309
PB  - Science Publishing Group
SN  - 2639-9733
UR  - https://doi.org/10.11648/j.ajai.20250902.29
AB  - The convergence of Artificial Intelligence (AI) with DevOps, DataOps, and MLOps has transformed the software development lifecycle, enabling scalable, automated, and intelligent systems. This paper explores the transition from traditional DevOps to MLOps, emphasizing the integration of machine learning workflows into continuous integration, deployment, and training pipelines. We present a practical framework for implementing MLOps using tools such as MLflow, Airflow, and Kubernetes, and address challenges like overfitting, underfitting, and model drift. The proposed architecture leverages Docker and ONNX for model packaging and deployment, ensuring reproducibility and cross-platform compatibility. Through real-world examples and pipeline automation strategies, we demonstrate how MLOps enhances model reliability, governance, and performance monitoring in dynamic environments. This study contributes to the growing body of knowledge on AI-driven DevOps by offering actionable insights for researchers and practitioners aiming to build robust ML systems. Build an Apache Airflow pipeline to load, train, and evaluate a ML model, store it, and use it for inferencing by deploying the model with a sleek Streamlit UI, Docker, and auto-scale it with Kubernetes as container orchestration tool. Techniques for implementing and automating continuous integration (CI), continuous delivery (CD), and continuous training (CT) for machine learning (ML) systems. This document applies primarily to predictive AI systems.
VL  - 9
IS  - 2
ER  -

Copy | Download