2022 to 2023: MLOps
Two years spent learning to ship models, with CI/CD, DVC, MLflow, YOLO object detection in production, SageMaker endpoints and the first private client work.
- Status
- stable
- Trust
- human-reviewedhuman:rrizwan98 ·
3 min read
- Approved
- human:rrizwan98 ·
A model in a notebook is a science project. These two years are about shipping them. The repositories change from single models to pipelines, and for the first time some of the work is private: 3 of 14 repositories in 2022, 6 of 23 in 2023.1
What happened
- MLOps pipelines. CI/CD, DVC, MLflow, Dagshub, Git workflows and end-to-end project structure, from an iris dataset in April 2022 to a full MLOps portfolio in June 2022, then end-to-end projects with MLflow and deployment through 2023.2
- Computer vision in production. YOLOv5 fire-and-smoke detection in 2022; YOLOv5, YOLOv7 and YOLOv8 object detection with MLOps in 2023; kidney-disease and chest-infection classification apps.
- AWS at depth. SageMaker training and endpoints, an endpoint wired to a Lambda function, a Hugging Face model deployed on SageMaker, a serverless SaaS API.
- Healthcare AI. A chatbot API and a cancer-care chatbot API built for real users, both private, in early 2022.2
- Sideways again. A Next.js repository in April 2023, and Web3 and metaverse assignments in 2022, alongside forks of the Panaverse curriculum, a Solidity smart-contracts course and the SageMaker book.
The public repositories
| Created | Repository | What it is |
|---|---|---|
| 2022-01-05 | Serverless-Api | Serverless SaaS API on AWS |
| 2022-01-31 | Dr.-Dinish- | Project repository |
| 2022-04-07 | MLOps | MLOps exercises |
| 2022-04-25 | iris-data-using-MLOps | First MLOps pipeline on the iris dataset |
| 2022-04-28 | web-Images-scraping | Image scraping for datasets |
| 2022-06-21 | MLOps-Portfolio | CI/CD pipelines, DVC, MLflow, Git workflow and Heroku deployment in one structure |
| 2022-08-03 | Data-Science-portfolio | Machine learning, deep learning and SageMaker work in one place |
| 2022-08-11 | fire_and_smoke_detection | Fire and smoke detection |
| 2022-08-14 | Raza_portfolio | Data science and MLOps portfolio |
| 2022-09-12 | learn-Sagemaker | SageMaker learning |
| 2022-09-26 | web3-and-metaverse-assignments | Course assignments; also holds a YOLOv5 fire-detection exercise |
| 2023-04-18 | nextjs | Next.js learning |
| 2023-05-03 | end_to_end_MLOps_projects | End-to-end MLOps projects |
| 2023-05-17 | MLOps-apply-on-BTC-Data | MLOps on Bitcoin price data |
| 2023-06-07 | Apply-Mlops-DVC-concepts-on-image-dataset | DVC on an image dataset |
| 2023-06-07 | datasets | Dataset store |
| 2023-07-21 | End_To_End_Cell_Object_Detection_Using_Yolo_V8_With_Deployment | YOLOv8 cell detection, trained and deployed with MLOps practices |
| 2023-07-31 | End_to_End_ML_Project_with_mlflow_and_Deployment | End-to-end project with MLflow and DVC, deployed |
| 2023-09-22 | Image-Classification-Sagemaker-with-MLOps | Image classification on SageMaker with MLOps |
| 2023-09-23 | End-to-End-Chest-Infection-Classification-App-With-MLOps | Chest-infection classification app |
| 2023-09-23 | yolov5-object-detection-with-MLOps | YOLOv5 detection with MLOps |
| 2023-09-23 | models | Model store |
| 2023-10-04 | Image-Classification-Model-Sagemaker-Connect-To-Lambda-Function | SageMaker endpoint connected to a Lambda function |
| 2023-10-07 | Kidney-Disease-Classification-Deep-Learning-Project | Kidney-disease classification |
| 2023-10-23 | Deploy-hugging-face-Model-on-AWS-sagemaker | Hugging Face model on SageMaker |
The three December 2023 repositories that close the year belong to the next page, because they mark the turn.
Private work
- 2022, three private repositories: two chatbot APIs, one of them for a cancer-care service, and an MLOps-on-SageMaker build.
- 2023, six private repositories: an object-detection job on YOLOv7, an MLOps exercise on a neural network, an image-scraping tool, a data pipeline, and in December the first two generative-AI repositories, a Gemini experiment and an agriculture solution.1
Private repositories are counted here and never named.
What this era taught
Deployment, versioning and monitoring are not extras. They are the product.2
Next: December 2023: the generative turn.
Footnotes
Sources
- [^github-repos]GitHub API, repositories owned by rrizwan98, read 26 August 2026
- [^journey-note]From Predictive AI to Forward Deployed Engineer: My 6-Year Journey, Raza Rizwan, July 2026