Published inITNEXT·PinnedMember-onlyPrinciples, Patterns, and Practices for Effective Infrastructure as CodeDeliver Infrastructure and Software running on it Rapidly and Reliably at Scale — Table of contents What is Infrastructure as Code Key Principles - Idempotency - Immutability Patterns and Practices - Everything in Source Control - Modularize and Version - Documentation - Testing - Security and Compliance - Automate Execution from a Shared Environment — Infrastructure as Code Pipeline — GitOps ConclusionDevOps11 min readDevOps11 min read
Published inITNEXT·Nov 17, 2020Member-onlyPlatform Engineering: Challenges and SolutionsBased on my experience leading Platform Engineering initiative at various organizations — I talked about what Platform Engineering(PE) is, when is it useful, and the challenges I have seen working with it in my previous article. In this article, we will go through solutions that have worked for my teams and me in resolving those challenges. …Automation11 min readAutomation11 min read
Published inITNEXT·Oct 15, 2020Member-onlyPlatform Engineering: Using it to Gain Competitive AdvantageWhat is it, When is it useful, and Challenges with it — Table of contents What is Platform Engineering? Platform Engineering Team Who are the customers? When is it useful? Challenges with Platform Engineering Conclusion Caution: Platform is a widely used term to define various types of platforms, so be careful about using it. At one place, we chose not to use the term platform…Platform Engineering6 min readPlatform Engineering6 min read
Published inTowards Data Science·Sep 23, 2020Member-onlyContinuous Delivery for Machine Learning SystemsDeploying Machine Learning Systems to Production safely and quickly in a sustainable way — Table of contents What is Continuous Delivery? Continuous Integration Continuous Delivery vs. Continuous Deployment Machine Learning Workflow How does Continuous Delivery help with ML challenges? Data Management - Automated Data Pipeline Experimentation - Training Code - Training Process Production Deployment - Application Code Bringing it all together People ConclusionMachine Learning9 min readMachine Learning9 min read
Published inTowards Data Science·Jun 21, 2020Member-onlyChallenges Deploying Machine Learning Models to ProductionMLOps: DevOps for Machine Learning — Table of contents Traditional Software Development vs Machine Learning Machine Learning Workflow Stage #1: Data Management - Large Data Size - High Quality - Data Versioning - Location - Security & Compliance Stage #2: Experimentation - Constant Research and Experimentation Workflow - Tracking Experiments - Code Quality - Training Time & Troubleshooting - Model Accuracy Evaluation - Retraining - Infrastructure Requirements Stage #3: Production Deployment - Offline/Online Prediction - Model Degradation Conclusion…Machine Learning7 min readMachine Learning7 min read
