👋 Hi! I'm Bibin Wilson. In each TL;DR edition, I share practical DevOps, Kubernetes, Cloud, and MLOps resources to help you stay up to date. Subscribe here to get every TL;DR edition in your inbox.
This Week in DevOpsCube ☕
How Razorpay Secured 7,000+ Kubernetes Nodes
Kubeflow for MLOPs: A Practical Crash Course
CRI vs CNI vs CSI vs CDI in Kubernetes
How is MLOPS different from usual DevOps workflows
DevOps Remote Jobs
and more..
☕ Grab a coffee and catch up on this week’s DevOps, MLOps, and AI insights and resources.
Sponsored by Cloudways
Simplify Cloud Hosting with Cloudways: Spend less time managing infrastructure and more time shipping. Cloudways handles server setup, security patches, and scaling on top providers like AWS and DigitalOcean. Start Your Free Trial.
📢 Latest in DevOps
Amazon S3 Adds Object Annotations: Lets you attach large, searchable notes directly to your files, replacing old methods that only allowed tiny, unchangeable labels.
GitLost Exposes GitHub AI Workflow Risk: Noma Security discovered GitLost, a prompt injection attack that can trick GitHub Agentic Workflows into leaking private repository data through a public GitHub issue.
AWS DevOps Agent Adds Release Management: AWS DevOps Agent can now automatically review code changes, detect release risks, and run release tests before deployment to help teams ship safer software.
Harness Launches Autonomous Worker Agents: Harness introduced Autonomous Worker Agents that can deploy, test, and scan applications while operating within existing CI/CD pipelines and enterprise governance controls.
AlloyDB AI Functions Get Faster and Cheaper: Google Cloud enhanced AlloyDB AI Functions with optimized local AI execution, delivering faster queries, lower inference costs, and new built-in AI functions like sentiment analysis and summarization.
🎯 How is MLOps Different From Usual DevOps Workflows?
MLOps is different from the usual DevOps workflows, just like machine learning is different from traditional software development.
For example, in the SDLC we manage the lifecycle of code. MDLC adds two more dimensions. ie, data and models. So, in MLOps, you need to track three components: code, data, and models
While in traditional DevOps, you only track code. It is why the system is more complex to deploy and maintain.
If you are a DevOps engineer looking to upskill in MLOps workflows,
We have written a detailed guide where we share learnings on building MLOps skills as a DevOps engineer.
👉 Detailed Roadmap: DevOps to MLOps
🤖 Kubeflow for MLOPs: A Practical Crash Course
In this guide, we covered the following.
Understand the Kubeflow stack
Learn how Kubeflow Pipelines work
Explore the KFP architecture and execution flow
Creating a KFP pipeline from scratch on Kubernetes
A look into Cloudflare’s in-house MLOPs platform, and more
👉Start Here: Kubeflow for MLOps
📖 Worth Reading This Week
Chaos Engineering at Flipkart: Learn how Flipkart built a multi-tenant chaos engineering platform with LitmusChaos to improve Kubernetes resilience before high-traffic sales events.
Scaling 1 Million Lambda Functions: Learn how ProGlove scaled to 1 million+ AWS Lambda functions, with practical lessons on multi-account architecture, observability, cost optimization, and operating serverless workloads at massive scale.
How Razorpay Secured 7,000+ Kubernetes Nodes: Learn how Razorpay automated RBI compliance across 7,000+ Kubernetes nodes using Policy as Code, enabling secure and compliant operations at fintech scale.
Cloudflare's Unified Data Platform: Learn how Cloudflare built Town Lake, a unified data platform, and Skipper, an AI agent that lets employees query company data using natural language.
Argo CD 2026 User Survey Results: Explore how organizations are using Argo CD for GitOps, including adoption trends, deployment practices, and production usage from the 2026 community survey.
⚙️ MLOps: Airflow + DVC Pipeline in Kubernetes
In this guide, you will learn about:
Airflow and DAGs explained for DevOps engineers
How Airflow, DVC, EKS, and S3 work together in a real DVC pipeline DAG
Deploying the pipeline on Airflow (EKS)
Verifying data versioning
👉 Read it Here: Airflow + DVC Pipeline in K8s
📋 Changelog: Releases & Breaking Changes
GitLab 19.1 Released: GitLab 19.1 adds unified SARIF vulnerability reporting, AI security improvements, stronger AI agent controls, and enhanced DevSecOps capabilities.
etcd 3.7 Released: etcd 3.7 introduces the new RangeStream API, improves performance and reliability, removes the legacy v2 store, and modernizes its protobuf implementation.
Docker Desktop Release Notes: Docker Desktop security fix including a VM panic bug, a container-to-host escape in the Model Runner MLX backend, and an unprivileged-to-root escape via page cache writes.
KServe 0.19.0 Released: KServe 0.19.0 brings new capabilities and improvements for serving generative and predictive AI workloads on Kubernetes, along with bug fixes and platform enhancements.
Azure DevOps WIF Issuer Retirement: Microsoft is retiring the Azure DevOps Workload Identity Federation issuer on July 1, 2027, so affected service connections must be migrated to the Microsoft Entra issuer.
🛠️ DevOps Tool of the Week (Inspektor-Gadget)
Inspektor Gadget is an open-source framework that uses eBPF to inspect and debug Kubernetes clusters and Linux hosts.
It automatically identifies which Pod, container, and namespace each kernel event belongs to.
Here is what it does👇
- Trace DNS lookups, TCP connections, file opens, and process execs across your workloads in real time.
- Capture packets inside any container using tcpdump-style pcap filters.
- Spot OOM kills, dropped TCP packets, and retransmissions the moment they happen.
- Filter events by pod or container name, with the matching done in-kernel to keep overhead low.
- Generate seccomp profiles and NetworkPolicies from what your workloads actually do at runtime.
👉 Start Here: Inspektor Gadget
🎓 Complete Kubernetes & CKA Course
10,000+ engineers have learned through DevOpsCube courses.
From container fundamentals to CKA preparation, every course is self-paced and built around real scenarios. This is not a long video lecture series.
The CKA course is text-based, illustration-rich, and designed for faster learning and quick revision whenever you need it.
👉 Coupon: Use code FLASH40 to get 40% OFF today.
Note: This is a community only offer! Make use of it!
🛡️ CRI Vs CNI Vs CSI Vs CDI in Kubernetes
CRI (Container Runtime Interface): How kubelet talks to container runtimes
CNI (Container Network Interface): How pods get network connectivity
CSI (Container Storage Interface): How pods get persistent storage
CDI (Container Device Interface): How pods access
hardware devices (NVIDIA, AMD, Intel)
Here, the key pattern is, K8s defines the interface, vendors build the plugins.
💼 DevOps Remote Jobs
Crop.photo by Evolphin - Senior Platform Engineer(DevOps) (5-9 Yrs)
Miratech - AWS DevOps Engineer (4+ Yrs)
Mactores - AWS DevOps Engineer(Associate) (2+ Yrs)
Kayzen - DevOps Engineer (2 Yrs)
MNJ Software - DevOps Monitoring Engineer (3-6 Yrs)
Provenir - Associate Cloud Operations Engineer (3-5 Yrs)
HPE - DevOps Engineer (5-8 Yrs)
Portainer.io - Platform Engineer (3+ Yrs)
HubSpot - DevOps Engineer (2+ Yrs)
SentinelOne - Senior DevOps Engineer
Subscribe to Join DevOps & MLOps Community
20,000+ trusted Engineers read our newsletter, where we simplify DevOps, MLOps and Kubernetes into practical, easy-to-follow guides.
DevOpsCube team:
Lead Editor: Bibin Wilson / Designer: Aljin / Tech Researcher: Aswin
Interested in advertising with us? Get in touch
Need DevOps consulting? Contact us
Want to learn from courses? Check our courses
Forwarded this email? Join our mailing list
