Interview Query

Image Classification Pipeline

Start Timer

0:00:00

Upvote
0
Downvote
Save question
Mark as completed
View comments

You work as a machine learning engineer at a cloud-based SaaS company. The company wants to automate the process of building, training, and deploying image classification models using datasets stored in S3. The models must be deployed on Kubernetes (EKS), and the entire workflow should be automated with a CI/CD pipeline.

How would you design this end-to-end system?

.
.
.
.
.
Topic
ML Ops & Training Pipelines
Tags
Kubernetes
S3
Prometheus
Grafana
Data Pipeline
Docker Containerization
Amazon EKS
Containerization
FastAPI Framework
GitHub Actions
Jenkins CI/CD
Data Validation
System Monitoring
Test Automation
Classification Algorithms
Blue Green Deployment
Terraform
Canary Deployment
Checkpointing in Computing
Cloud Computing
Data Augmentation Techniques
Data Preprocessing
Experiment Tracking Software
Model Explainability
Kubernetes
Machine Learning Model Training
Performance Metrics
AWS Secrets Manager
Hyperparameter Tuning
Fluentd
MLflow
Kubeflow Pipelines
Metaflow
Kubeflow Metadata
ONNX
TorchScript
TensorFlow Serving
TorchServe
Distributed Training
RBAC
Early Stopping
Model Packaging
CI/CD Automation
Category
System Design
Position
Data Engineer
Difficulty
Hard
Success
N/A
Last asked
This question was asked on July 16, 2025
Companies
What did you think?


Comments