Skip to content

Commit eaf5c26

Browse files
alexeygrigorevalexeygrigorev
committed
Add MLOps Zoomcamp documentation section
Add a course documentation section for the MLOps Zoomcamp under courses/mlops-zoomcamp/, mirroring the ML Zoomcamp page set: index, prerequisites, getting-started, curriculum, environment-setup, whats-new, project, resources. MLOps previously had no docs page of its own; only the shared zoomcamp-logistics pages covered it. nav_order 5 (after LLM).
1 parent 4ee4306 commit eaf5c26

8 files changed

Lines changed: 326 additions & 0 deletions

File tree

‎courses/mlops-zoomcamp/curriculum.md‎

Lines changed: 61 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,61 @@
1+
---
2+
title: "Curriculum"
3+
layout: default
4+
nav_order: 3
5+
parent: MLOps Zoomcamp
6+
has_children: false
7+
---
8+
9+
# Curriculum
10+
11+
The MLOps Zoomcamp covers six main modules plus a final project. Each module has video lectures, hands-on material, and a homework assignment.
12+
13+
For the canonical curriculum (videos, code, exact homework questions), see the [GitHub repository](https://github.com/DataTalksClub/mlops-zoomcamp).
14+
15+
## Modules
16+
17+
[Module 1: Introduction](https://github.com/DataTalksClub/mlops-zoomcamp/tree/main/01-intro)
18+
19+
- What is MLOps and why it matters.
20+
- MLOps maturity model.
21+
- The NY Taxi dataset used as the running example.
22+
- Course structure and environment setup.
23+
24+
[Module 2: Experiment Tracking & Model Management](https://github.com/DataTalksClub/mlops-zoomcamp/tree/main/02-experiment-tracking)
25+
26+
- Experiment tracking with MLflow.
27+
- Saving and loading models.
28+
- The model registry.
29+
30+
[Module 3: Orchestration & ML Pipelines](https://github.com/DataTalksClub/mlops-zoomcamp/tree/main/03-orchestration)
31+
32+
- Turning notebooks into orchestrated ML pipelines.
33+
- Workflow orchestration.
34+
35+
[Module 4: Model Deployment](https://github.com/DataTalksClub/mlops-zoomcamp/tree/main/04-deployment)
36+
37+
- Online vs. offline deployment.
38+
- Web service deployment with Flask.
39+
- Streaming deployment with AWS Kinesis and Lambda.
40+
- Batch scoring for offline processing.
41+
42+
[Module 5: Model Monitoring](https://github.com/DataTalksClub/mlops-zoomcamp/tree/main/05-monitoring)
43+
44+
- Monitoring ML services.
45+
- Web service monitoring with Prometheus, Evidently, and Grafana.
46+
- Batch job monitoring with Prefect, MongoDB, and Evidently.
47+
48+
[Module 6: Best Practices](https://github.com/DataTalksClub/mlops-zoomcamp/tree/main/06-best-practices)
49+
50+
- Unit and integration testing.
51+
- Linting, formatting, and pre-commit hooks.
52+
- CI/CD with GitHub Actions.
53+
- Infrastructure as Code with Terraform.
54+
55+
[Final Project](https://github.com/DataTalksClub/mlops-zoomcamp/tree/main/07-project)
56+
57+
- An end-to-end project that integrates experiment tracking, orchestration, deployment, and monitoring.
58+
59+
## Homework and project
60+
61+
Each module has a homework assignment. To earn the certificate, you also complete the final project during a live cohort. See [Project]({{ '/courses/mlops-zoomcamp/project/' | relative_url }}) for details.

‎courses/mlops-zoomcamp/environment-setup.md‎

Lines changed: 50 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,50 @@
1+
---
2+
title: "Environment Setup"
3+
layout: default
4+
nav_order: 4
5+
parent: MLOps Zoomcamp
6+
has_children: false
7+
---
8+
9+
# Environment Setup
10+
11+
The MLOps Zoomcamp uses Python with common ML and MLOps tooling: MLflow, Prefect, Docker, Evidently, Grafana, Prometheus, Terraform, and AWS for cloud examples. This page covers the high-level setup decisions. For step-by-step setup, follow the videos in [Module 1](https://github.com/DataTalksClub/mlops-zoomcamp/tree/main/01-intro).
12+
13+
## Where to run the course
14+
15+
You can work locally or on a cloud VM. Module 1 demonstrates both:
16+
17+
- A local machine with Python and Docker installed.
18+
- GitHub Codespaces.
19+
- A cloud VM (the course shows an AWS EC2 instance).
20+
21+
A cloud VM is convenient because the later modules use AWS services anyway, and it keeps your environment consistent.
22+
23+
## Python and dependency management
24+
25+
The course uses Python with the scientific stack (pandas, scikit-learn) plus MLOps libraries. You can manage dependencies with `uv`, conda, or pip + venv. The choice does not affect the course content.
26+
27+
On Windows, Anaconda is often the easiest way to get Python plus scientific libraries running. Docker Desktop covers the container side.
28+
29+
## Docker
30+
31+
Docker is used throughout the course for packaging models, running services, and reproducing environments. Install Docker and Docker Compose early. Module 1 includes the installation steps.
32+
33+
## Cloud (AWS)
34+
35+
The course uses AWS for several modules:
36+
37+
- Module 4 deploys a streaming service with Kinesis and Lambda.
38+
- Module 6 provisions infrastructure with Terraform.
39+
40+
New AWS accounts get free tier credits. You introduce each service as the modules reach it, so you do not need prior AWS experience.
41+
42+
## Tools introduced per module
43+
44+
- Module 2: MLflow.
45+
- Module 3: a workflow orchestrator.
46+
- Module 4: Flask, AWS Kinesis, AWS Lambda, Docker.
47+
- Module 5: Evidently, Grafana, Prometheus, Prefect, MongoDB.
48+
- Module 6: pytest, pre-commit, GitHub Actions, Terraform.
49+
50+
Each module README lists exactly what to install for that module.

‎courses/mlops-zoomcamp/getting-started.md‎

Lines changed: 44 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,44 @@
1+
---
2+
title: "Getting Started"
3+
layout: default
4+
nav_order: 2
5+
parent: MLOps Zoomcamp
6+
has_children: false
7+
---
8+
9+
# Getting Started
10+
11+
For the cross-course onboarding, read [Joining a Cohort]({{ '/courses/zoomcamp-logistics/joining/' | relative_url }}) first. It covers registration, account setup, calendar, and newsletter updates.
12+
13+
This page covers the MLOps Zoomcamp specifics.
14+
15+
## Star the GitHub repository
16+
17+
[github.com/DataTalksClub/mlops-zoomcamp](https://github.com/DataTalksClub/mlops-zoomcamp)
18+
19+
Star it so you can find it later. All course materials are here, with each module having its own folder. Cohort-specific homework and deadlines are under the `cohorts/` folder.
20+
21+
See [Resources]({{ '/courses/mlops-zoomcamp/resources/' | relative_url }}) for more.
22+
23+
## Join the MLOps Zoomcamp Slack channel
24+
25+
After joining the [DataTalks.Club Slack workspace](https://datatalks.club/slack), find the `#course-mlops-zoomcamp` channel. This is where questions, announcements, and discussion happen.
26+
27+
Before asking questions, read [Asking Questions]({{ '/courses/zoomcamp-logistics/asking-questions/' | relative_url }}) and the [community guidelines](https://datatalks.club/slack/guidelines.html).
28+
29+
## Watch the course videos
30+
31+
All lectures are on the [course YouTube playlist](https://www.youtube.com/playlist?list=PL3MmuxUbc_hIUISrluw_A7wDSmfOhErJK). The repository README and each module folder link to the relevant videos in order.
32+
33+
## Set up your environment
34+
35+
Module 1 walks through environment setup, including Python, Docker, and an optional cloud VM. See [Environment Setup]({{ '/courses/mlops-zoomcamp/environment-setup/' | relative_url }}) for the high-level decisions before you start.
36+
37+
## How to work through the course
38+
39+
1. Start in the module folder you are working on.
40+
2. Read the README in that folder for an overview.
41+
3. Follow the links to the video lectures.
42+
4. Complete the homework assignment.
43+
5. Check the cohort folder for cohort-specific materials and deadlines.
44+
6. Finish with the final project, which ties the modules together.

‎courses/mlops-zoomcamp/index.md‎

Lines changed: 27 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,27 @@
1+
---
2+
title: "MLOps Zoomcamp"
3+
layout: default
4+
nav_order: 5
5+
parent: Courses
6+
description: "Free, hands-on MLOps course on productionizing ML services"
7+
has_children: true
8+
---
9+
10+
# MLOps Zoomcamp
11+
12+
Welcome to the MLOps Zoomcamp. It is a completely free, hands-on course on productionizing machine learning services, from training and experimentation to deployment and monitoring.
13+
14+
## Docs order
15+
16+
Read in this order:
17+
18+
1. [Community Guidelines]({{ '/general/guidelines/' | relative_url }}) - code of conduct, how to ask questions, how to use Slack channels, how to promote your work.
19+
2. [Zoomcamp Logistics]({{ '/courses/zoomcamp-logistics/' | relative_url }}) - how DataTalks.Club zoomcamps work in general (cohort schedule, joining, live sessions, homework, project, peer review, certification). Most of your logistical questions are answered there.
20+
3. The pages in this section - what is specific to the MLOps Zoomcamp (curriculum, environment, project, MLOps-specific resources).
21+
4. The [MLOps Zoomcamp FAQ](https://datatalks.club/faq/mlops-zoomcamp.html) - module-specific and technical questions from previous cohorts.
22+
23+
For platform mechanics (where to click on the submission form, how the leaderboard appears), see [Course Management Platform]({{ '/courses/course-management-platform/' | relative_url }}).
24+
25+
## Current status
26+
27+
The MLOps Zoomcamp is currently available for self-paced study. We don't plan to run a live cohort in 2026. [Register here](https://airtable.com/shrCb8y6eTbPKwSTL) if you'd like to be notified in case we run it again.

‎courses/mlops-zoomcamp/prerequisites.md‎

Lines changed: 35 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,35 @@
1+
---
2+
title: "Prerequisites"
3+
layout: default
4+
nav_order: 1
5+
parent: MLOps Zoomcamp
6+
has_children: false
7+
---
8+
9+
# Prerequisites
10+
11+
The MLOps Zoomcamp builds on existing machine learning and programming experience. It is not a beginner ML course; it teaches you how to take models you already know how to build and run them in production.
12+
13+
For general expectations about zoomcamp time commitment, see [Before You Start]({{ '/courses/zoomcamp-logistics/before-you-start/' | relative_url }}).
14+
15+
## Required skills
16+
17+
You need these basics:
18+
19+
- Python programming.
20+
- Comfort with the command line.
21+
- Docker basics.
22+
- Machine learning fundamentals: training a model, evaluating it, and understanding the typical ML workflow. The [Machine Learning Zoomcamp](https://github.com/DataTalksClub/machine-learning-zoomcamp) is a good way to cover this.
23+
- At least 1 year of programming experience.
24+
25+
## You do not need
26+
27+
You do not need these before starting:
28+
29+
- Prior MLOps or DevOps experience. The course introduces experiment tracking, orchestration, deployment, and monitoring from the ground up.
30+
- Prior cloud experience. The course uses AWS but introduces the services it relies on.
31+
- Deep math. The focus is on engineering and operations, not modeling theory.
32+
33+
## If you are missing a prerequisite
34+
35+
If you are new to machine learning, take the [Machine Learning Zoomcamp](https://github.com/DataTalksClub/machine-learning-zoomcamp) first. If you are comfortable building models but new to Python tooling, Docker, or the command line, work through Module 1, which covers environment setup before the MLOps content begins.

‎courses/mlops-zoomcamp/project.md‎

Lines changed: 37 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,37 @@
1+
---
2+
title: "Project"
3+
layout: default
4+
nav_order: 6
5+
parent: MLOps Zoomcamp
6+
has_children: false
7+
---
8+
9+
# Project
10+
11+
For the cross-course logistics, see [Final Project (Zoomcamp Logistics)]({{ '/courses/zoomcamp-logistics/project/' | relative_url }}). That page covers attempts, deadlines, peer review, and certification mechanics.
12+
13+
This page covers what is specific to the MLOps Zoomcamp.
14+
15+
## Goal
16+
17+
The project is an end-to-end MLOps pipeline. You pick a problem and a dataset, train a model, and then apply the practices from the course to put it into production and operate it.
18+
19+
## What the project should demonstrate
20+
21+
A complete project applies the modules together:
22+
23+
- A clear problem description.
24+
- Experiment tracking and a model registry (Module 2).
25+
- A training pipeline that can be run reproducibly (Module 3).
26+
- Model deployment, as a web service, streaming, or batch (Module 4).
27+
- Model monitoring (Module 5).
28+
- Best practices: tests, linting, CI/CD, and Infrastructure as Code where appropriate (Module 6).
29+
- Reproducibility: clear instructions and pinned dependencies so a reviewer can run it.
30+
31+
## Where to find the details
32+
33+
The [07-project folder](https://github.com/DataTalksClub/mlops-zoomcamp/tree/main/07-project) in the repository contains the project instructions and the evaluation criteria. The rubric there is the authoritative description of what is graded.
34+
35+
## Certificate
36+
37+
Completing the project during a live cohort is how you earn the certificate. Since there is no live 2026 cohort, the project is currently for self-assessment and portfolio building. See [Certification]({{ '/courses/zoomcamp-logistics/certification/' | relative_url }}) for how certification works in general.

‎courses/mlops-zoomcamp/resources.md‎

Lines changed: 53 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,53 @@
1+
---
2+
title: "Resources"
3+
layout: default
4+
nav_order: 7
5+
parent: MLOps Zoomcamp
6+
has_children: false
7+
---
8+
9+
# Resources
10+
11+
Course-specific links for the MLOps Zoomcamp. For general zoomcamp logistics, see [Zoomcamp Logistics]({{ '/courses/zoomcamp-logistics/' | relative_url }}).
12+
13+
## GitHub Repository
14+
15+
The repository is your primary navigation tool throughout the course.
16+
17+
[github.com/DataTalksClub/mlops-zoomcamp](https://github.com/DataTalksClub/mlops-zoomcamp)
18+
19+
How to use it:
20+
21+
1. Start in the module folder you are working on.
22+
2. Read the README in that folder for an overview.
23+
3. Follow the links to video lectures.
24+
4. Complete the homework assignment.
25+
5. Check the cohort folder for cohort-specific materials.
26+
27+
## Video lectures
28+
29+
All lectures are on the [course YouTube playlist](https://www.youtube.com/playlist?list=PL3MmuxUbc_hIUISrluw_A7wDSmfOhErJK).
30+
31+
## FAQ
32+
33+
The [MLOps Zoomcamp FAQ](https://datatalks.club/faq/mlops-zoomcamp.html) collects module-specific and technical questions from previous cohorts.
34+
35+
## Slack
36+
37+
Questions and discussion happen in the `#course-mlops-zoomcamp` channel on [DataTalks.Club Slack](https://datatalks.club/slack.html). Read [Asking Questions]({{ '/courses/zoomcamp-logistics/asking-questions/' | relative_url }}) first.
38+
39+
## Tools used in the course
40+
41+
- [MLflow](https://mlflow.org/) - experiment tracking and model registry.
42+
- [Prefect](https://www.prefect.io/) - workflow orchestration.
43+
- [Evidently](https://www.evidentlyai.com/) - data and model monitoring.
44+
- [Grafana](https://grafana.com/) and [Prometheus](https://prometheus.io/) - metrics and dashboards.
45+
- [Docker](https://www.docker.com/) - containerization.
46+
- [Terraform](https://www.terraform.io/) - Infrastructure as Code.
47+
- [AWS](https://aws.amazon.com/) - cloud deployment (Kinesis, Lambda).
48+
49+
## Instructors
50+
51+
- [Cristian Martinez](https://www.linkedin.com/in/cristian-javier-martinez-09bb7031/)
52+
- [Alexey Grigorev](https://www.linkedin.com/in/agrigorev/)
53+
- [Emeli Dral](https://www.linkedin.com/in/emelidral/)

‎courses/mlops-zoomcamp/whats-new.md‎

Lines changed: 19 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,19 @@
1+
---
2+
title: "What's New"
3+
layout: default
4+
nav_order: 5
5+
parent: MLOps Zoomcamp
6+
has_children: false
7+
---
8+
9+
# Changes
10+
11+
Notable points about the current state of the course.
12+
13+
## Current status
14+
15+
The MLOps Zoomcamp is in self-paced mode. We don't plan to run a live cohort in 2026. All materials remain freely available, and the GitHub repository is the source of truth for the latest content.
16+
17+
For the canonical list of changes, follow the commit history and module READMEs in the [GitHub repository](https://github.com/DataTalksClub/mlops-zoomcamp). Materials from previous cohorts are preserved under the `cohorts/` folder.
18+
19+
For the full curriculum, see [Curriculum]({{ '/courses/mlops-zoomcamp/curriculum/' | relative_url }}). For environment setup, see [Environment Setup]({{ '/courses/mlops-zoomcamp/environment-setup/' | relative_url }}).

0 commit comments

Comments
 (0)