Hands-On MLOps Bootcamp Duration: 5 Days | Session Time: 2 Hours per Day Mode: Fully Lab-Based | Cloud-Ready | GitHub Access Provided This specific 5-day MLOps Bootcamp is tailored for students and working professionals who want to master the technical essentials of ML deployment, CI/CD, experiment tracking, and containerization — all within 10 focused hours. No fluff. Just deeply practical and job-relevant knowledge to make you interview- and project-ready. Organized by: Mindbox Trainings Format: Intensive 5-Day Technical Bootcamp Schedule: May 19th to 23rd, 2025 | Time: 7:00 AM – 9:00 AM IST Mode: Online | Fully Hands-On Labs | Cloud + GitHub Enabled Audience: Final-year students, job seekers, early professionals. Prerequisites: Laptop, GitHub access, Free or Trial GCP Account 🗓️ 5-Day MLOps Curriculum – Detailed Breakdown 📅 Day 1: Foundations of ML & MLOps Topics & Activities: Real-world problems in ML project delivery Why MLOps is essential today Core concepts: data, model, features, parameters Demo: Build a basic ML model to predict apartment prices Visualize training and evaluation using matplotlib Outcomes: ✅ Understand the ML lifecycle and challenges ✅ Build and evaluate an ML model ✅ Interpret model behavior using charts ................................................................................................................................. 📅 Day 2: Experiment Tracking with MLflow Topics & Activities: Track experiments and models using MLflow Concepts: runs, parameters, metrics, artifacts Explore MLflow UI for comparisons Demo: Track and visualize multiple training runs Outcomes: ✅ Track every model version like a pro ✅ Use MLflow UI to compare models ✅ Integrate tracking into your ML scripts ................................................................................................................................. 📅 Day 3: Model Deployment - Local & Cloud Topics & Activities: Serve models as APIs Test API locally with Postman/curl Demo: Deploy the model on a GCP VM and access it remotely Automate model startup with systemd or screen Outcomes: ✅ Expose your model as a REST API ✅ Deploy it on a cloud VM ✅ Enable real-time access and testing ................................................................................................................................. 📅 Day 4: Docker + Data Versioning with DVC Topics & Activities: Part 1 - Docker: Containerize your ML app using Docker Optimize builds using caching Push/pull images via DockerHub Demo: Deploy containerized model in cloud Part 2 - Data Versioning: Why data versioning is critical Use Git + DVC for syncing datasets Remote storage setup (e.g., GCS bucket) Demo: Track data versions, sync across machines Outcomes: ✅ Build and deploy Dockerized ML apps ✅ Manage training data like code ✅ Collaborate across teams with synced code + data ................................................................................................................................. 📅 Day 5: CI/CD Pipelines for ML Projects Topics & Activities: What is CI/CD in MLOps? Define software + pipeline requirements Overview of GitHub Actions Create secrets to connect with GCP, DockerHub, etc. Demo: Build full ML pipeline using GitHub Actions Code checkout Docker build & push Cloud deployment API test post-deploy Outcomes: ✅ Automate your full ML pipeline ✅ Use GitHub Actions to build/test/deploy ✅ Deploy models with zero manual effort ................................................................................................................................. 💡 Lab Guarantee: Each session ends with a hands-on, replicable lab using: A laptop Cloud account (preferably GCP) Access to a GitHub repo (provided) 🎓 What You Take Away: Real-world end-to-end project experience Confidence in building MLOps-ready applications Full-stack ML understanding — from development to production Future-proof skills using Cloud + Open Source tools Recordings of each day’s session will be shared to ensure you can revisit the content anytime and reinforce your learning.