100-days-mlops-kodekloud

100 Days of MLOps Challenge [KodeKloud]

This repository contains the code and resources for the 100 Days of MLOps challenge by KodeKloud. The course is designed to help you learn and practice MLOps concepts and techniques over a span of 100 days.

---

Course Overview

This challenge walks through core MLOps practice end to end. You start with local Python setup, project structure, and code quality, then move through data versioning, experiment tracking, model registry, training pipelines, serving, monitoring, CI/CD, and Kubernetes-based deployment.

Prerequisite

Before starting, you should know basic Python, Git, command-line work, and simple machine learning workflows. Helpful extras: Docker, YAML, and comfort editing files in a terminal. I would recommend finishing this 100 days of DevOps challenge before starting here.

What You’ll Learn

By the end of the 100 days, you should be able to:

• Set up reproducible Python environments and standard ML project layouts
• Use DVC and MLflow for data, experiments, metrics, and model lifecycle tracking
• Package, test, lint, and automate ML workflows with modern Python tooling
• Build and containerize model serving APIs with Flask, FastAPI, BentoML, and Docker
• Add monitoring, drift detection, reporting, and retraining workflows
• Build CI/CD pipelines for model validation, registration, deployment, and rollback
• Orchestrate ML workloads on Kubernetes with Argo, Prefect, KServe, and Kubeflow Pipelines

Quick Start

1. Star this repository ⭐ to keep track of your progress
2. Fork the repo to create your own copy
3. Start with Day 001 - Create a Python Virtual Environment for ML
4. Follow the daily challenges in order
5. Share your progress using #100DaysOfMLOps

Want to start the official challenge? Use this KodeKloud link - it helps support this project!

Challenge Progress

Daily Challenges

Day	Challenge	Topics	Solution
01	Create a Python Virtual Environment for ML	Python, virtual environments, project setup	Solved
02	Set Up and Configure Jupyter Notebook Server	Jupyter, notebooks, local dev setup	Solved
03	Fix a Broken uv Lockfile Specification	uv, dependency locking, package resolution	Solved
04	Create a Standard ML Project Structure	repository structure, packaging, conventions	Solved
05	Create a Makefile for ML Workflow Automation	Makefile, task automation, developer workflow	Solved
06	Set Up Code Quality Tools for ML Code	linting, formatting, static analysis	Solved
07	Package an ML Project as Installable Python Package	packaging, setuptools, installation	Solved
08	Configure Pre-Commit Hooks for ML Repository	pre-commit, git hooks, code quality	Solved
09	Create a Custom ML Project Template with Cookiecutter	Cookiecutter, templates, scaffolding	Solved
10	Install and Initialize DVC in an ML Project	DVC, data versioning, initialization	Solved
11	Track a Dataset with DVC	DVC, dataset tracking, reproducibility	Solved
12	Configure a DVC Remote Storage	DVC, remote storage, artifact sync	Solved
13	Pull DVC-Tracked Data from Remote	DVC, pull, data restore	Solved
14	Create a DVC Pipeline for Data Processing	DVC pipelines, data processing, stages	Solved
15	Parameterize a DVC Pipeline	DVC, parameters, pipeline configuration	Solved
16	Track ML Metrics with DVC	DVC, metrics, experiment tracking	Solved
17	Run and Compare DVC Experiments	DVC, experiments, comparison	Solved
18	Version Datasets and Models Across Git Branches	Git branches, model versioning, datasets	Solved
19	Build Complete DVC ML Pipeline with Remote Storage and Experiments	DVC, remote storage, full pipeline	Solved
20	Install and Start the MLflow Tracking Server	MLflow, tracking server, setup	Solved
21	Log an ML Experiment to MLflow	MLflow, experiment logging, runs	Solved
22	Create and Organize MLflow Experiments	MLflow, experiment management, organization	Solved
23	Search and Query MLflow Runs	MLflow, search, querying runs	Pending
24	Enable MLflow Autologging	MLflow, autologging, instrumentation	Pending
25	Register, Version, and Manage Model Lifecycle	MLflow, model registry, lifecycle	Pending
26	Compare Model Runs and Select the Best	MLflow, model comparison, selection	Pending
27	Load Model from Registry with Custom Preprocessing	MLflow, registry, preprocessing	Pending
28	Fix a Broken MLflow Project and Re-Run It	MLflow, debugging, project recovery	Pending
29	Configure MLflow with Remote Tracking Server and Artifact Store	MLflow, remote tracking, artifacts	Pending
30	End-to-End MLflow Lifecycle: Train, Register, Serve, Monitor	MLflow, lifecycle, deployment, monitoring	Pending
31	Train a Scikit-Learn Model with Reproducible Script	scikit-learn, training script, reproducibility	Pending
32	Manage Training Configuration with YAML	YAML, configuration, training params	Pending
33	Evaluate a Trained Model and Generate Classification Report	evaluation, metrics, classification report	Pending
34	Implement Cross-Validation for Model Selection	cross-validation, model selection, evaluation	Pending
35	Hyperparameter Tuning with Optuna	Optuna, hyperparameter tuning, optimization	Pending
36	Automated Model Selection with FLAML AutoML	FLAML, AutoML, model selection	Pending
37	Distributed Model Training with Joblib Parallelization	joblib, parallelization, distributed training	Pending
38	Build Modular Training Pipeline with Config-Driven Stages	modular design, config-driven pipeline, training	Pending
39	Train a PyTorch Model with GPU Support and Checkpointing	PyTorch, GPU, checkpointing	Pending
40	Production Training System: Tracking, Tuning, and Model Selection	production training, tracking, tuning	Pending
41	Install and Initialize a Feast Feature Store	Feast, feature store, initialization	Pending
42	Define Feature Views in Feast	Feast, feature views, feature engineering	Pending
43	Materialize Features to the Online Store	Feast, online store, materialization	Pending
44	Store MLflow’s Admin Password in HashiCorp Vault	Vault, secrets, MLflow integration	Pending
45	Fix a Broken Vault KV Policy for the MLflow Reader	Vault, policies, access control	Pending
46	Author Data-Quality Expectations with Great Expectations	Great Expectations, data quality, expectations	Pending
47	Debug a Failing Great Expectations Checkpoint	Great Expectations, checkpoints, debugging	Pending
48	Publish Great Expectations Data Docs as a CI Artefact	Great Expectations, CI, documentation	Pending
49	Secrets + Data-Quality Integration Capstone	secrets management, data quality, capstone	Pending
50	Create Docker Image for ML Training Environment	Docker, containerization, training env	Pending
51	Create Multi-Stage Docker Build for ML Serving	Docker, multi-stage builds, serving	Pending
52	Set Up Local ML Dev Environment with Docker Compose	Docker Compose, local dev, services	Pending
53	Create GPU-Enabled Docker Image for Deep Learning	Docker, GPU, deep learning	Pending
54	Push ML Model Images to Container Registry	container registry, image push, Docker	Pending
55	Add Health Checks and Graceful Shutdown to ML Containers	health checks, shutdown, reliability	Pending
56	Automate ML Docker Image Building in CI Pipeline	CI, Docker build, automation	Pending
57	Serve an ML Model with Flask	Flask, serving, API	Pending
58	Serve an ML Model with FastAPI	FastAPI, serving, API	Pending
59	Run Batch Predictions on a Dataset	batch inference, predictions, dataset processing	Pending
60	Package a Model as a BentoML Service	BentoML, packaging, model service	Pending
61	Containerize an ML Model API with Docker	Docker, API containerization, serving	Pending
62	Implement A/B Testing for Model Deployment	A/B testing, deployment, experimentation	Pending
63	Implement Async Batch Prediction with Task Queue	async jobs, task queues, batch prediction	Pending
64	Serve Multiple Models Behind Unified API Gateway	API gateway, multi-model serving, routing	Pending
65	Implement Canary Deployment for Model Updates	canary deployment, rollout, release strategy	Pending
66	Production Model Serving with Docker Compose	Docker Compose, production serving, orchestration	Pending
67	Add Prometheus as a Grafana Data Source	Prometheus, Grafana, observability	Pending
68	Generate a Model Performance Report	model performance, reporting, metrics	Pending
69	Generate a Data Quality Report	data quality, reporting, validation	Pending
70	Create Automated Tests with Evidently Test Suites	Evidently, testing, monitoring	Pending
71	Set Up Evidently Monitoring Dashboard	Evidently, dashboard, monitoring	Pending
72	Set Up Drift Detection Alerts	drift detection, alerts, monitoring	Pending
73	Automatic Retraining Triggered by Drift Detection	drift detection, retraining, automation	Pending
74	Monitor Custom Business Metrics Alongside ML Metrics	business metrics, ML metrics, monitoring	Pending
75	End-to-End Monitoring: Prometheus, Grafana, Evidently	monitoring stack, observability, ML metrics	Pending
76	Create CI Pipeline for ML Code Linting and Testing	CI, linting, testing	Pending
77	Add Data Validation to CI Pipeline	CI, data validation, quality gates	Pending
78	Add Model Validation Tests to CI	CI, model validation, testing	Pending
79	Generate Model Performance Reports in CI with CML	CML, CI, reporting	Pending
80	Automate Model Registration in CI/CD	CI/CD, model registry, automation	Pending
81	Automate Model Deployment with CD Pipeline	CD, deployment, automation	Pending
82	End-to-End ML CI/CD Pipeline	CI/CD, automation, deployment	Pending
83	Automated Model Rollback with Health Checks	rollback, health checks, deployment safety	Pending
84	Production ML CI/CD with Multi-Environment Promotion	CI/CD, promotion, environments	Pending
85	Install Argo Workflows on Kubernetes	Argo Workflows, Kubernetes, installation	Pending
86	Create a Basic ML Training Workflow in Argo	Argo Workflows, training workflow, Kubernetes	Pending
87	Pass Data Between Argo Steps with Output Parameters and Branching	Argo, parameters, branching	Pending
88	Create an ML Pipeline with Prefect	Prefect, workflow orchestration, pipelines	Pending
89	Parallel Model Training with Argo withItems Fan-Out	Argo, parallelism, fan-out	Pending
90	Automated Retraining with Argo CronWorkflow	Argo, CronWorkflow, scheduled retraining	Pending
91	Production ML Pipeline: Argo Workflows + MLflow on Kubernetes	Argo, MLflow, Kubernetes pipeline	Pending
92	Deploy an ML Model on Kubernetes	Kubernetes, deployment, serving	Pending
93	Configure HPA for ML Serving Deployment	Kubernetes, HPA, autoscaling	Pending
94	Deploy a Model with KServe InferenceService	KServe, inference service, Kubernetes	Pending
95	Kubeflow Pipelines - Install and Run a Basic KFP Pipeline	Kubeflow Pipelines, installation, orchestration	Pending
96	GitOps Model Deployment with ArgoCD	GitOps, ArgoCD, deployment	Pending
97	Capstone (1/4): End-to-End ML System - Train, Register, Serve	capstone, training, registry, serving	Pending
98	Capstone (2/4): Monitoring and Automated Retraining	capstone, monitoring, retraining	Pending
99	Capstone (3/4): Orchestrate the Full MLOps Loop with Argo Workflows	capstone, orchestration, Argo Workflows	Pending
100	Capstone (4/4): Close the Loop with Prometheus + Grafana Observability	capstone, Prometheus, Grafana, observability	Pending

🤝 Contributing

Contributions are welcome! If you think any solution steps would be better, Here’s how you can help:

1. Fork the repository
2. Create a feature branch (git checkout -b feature/improvement-day-number)
3. Commit your changes (git commit -am 'Add new solution for days-number')
4. Push to the branch (git push origin feature/improvement-day-number)
5. Create a Pull Request

⭐ Show Your Support

If this repository helped you in your DevOps journey:

• Star this repository ⭐
• Fork it to create your own version
• Share it with fellow developers
• Follow me for more DevOps content

📄 License

This project is licensed under the GPL License - see the LICENSE file for details.

🔗 Connect With Me

• GitHub: @imshakil
• LinkedIn: Connect with me
• Medium: @imshakil

---

Happy Learning!

Remember: Start with Day 1 and build your skills progressively!

This site is open source. Improve this page.