The Kubernetes tool I always wished existed

I built my own Kubernetes IDE because existing ones suck, I’ve been working on Agentkube - an AI-native Kubernetes IDE that runs locally and it's light-weight. Built for Platform Engineers, SREs, Devops professionals and AI infra teams.

Think: Cursor for Kubernetes.

Available on macOS & Windows – and it’s free to use! 🎉

(Except AI features — I didn’t want to burn through credits too early 😅 but I’ll make sure everyone can try them soon.)



While it’s still solo-built (so expect a few rough edges), it’s real and live now! Here is the preview: https://www.youtube.com/watch?v=vdDqt7jYpsU

I’d love to hear from the DevOps community - especially those using Kubernetes or tried it

What are you using today? kubectl, Lens, k9s, Headlamp, Monokle, something else?

Any feedback is welcome - I’m trying to make Kubernetes more accessible, smart, and even enjoyable.

DM me if you liked something, feature requests, or bugs https://github.com/agentkube/agentkube/ \- or just say hi!

https://redd.it/1l2kgs1
@r_devops

I built a list of recent FAANG-style interview problems

I compiled a list from recent candidate reports, split between LC-original and non-LC interview questions.

Here’s what I found:

For LC-original questions that showed up in interviews, the most common tags were:
- Array
- Two Pointers
- Hash Map
- DP
- String
- Sorting

For questions that weren’t on LC (or were serious twists), the most common patterns were:
- Hash Map
- DP
- Greedy
- Sliding Window
- BFS / DFS
- String
- Memoization
- Heap

What surprised me was how often companies asked medium to hard problems that didn’t resemble anything in the standard prep sets. So I took some time to organized these questions with solution explanation as well.

Just sharing in case anyone else is trying to make sense of the prep landscape right now.

https://redd.it/1l2llfr
@r_devops

Phone screening questions

I have a 15 minute phone screening for a DevOps Engineer position in a couple of days. I have done my research on the interviewer and considering it's only 15 minutes, I can assume that it'll be behavioral questions.

What kind of questions could I expect is my question? Any suggestions would be greatly appreciated!

https://redd.it/1l2nxza
@r_devops

Building Production-Ready MySQL Infrastructure on GCP with OpenTofu/Terraform: A Complete Guide

As a Senior Solution Architect, I’ve witnessed the evolution of database deployment strategies from manual server configurations to fully automated infrastructure as code. Today, I’m sharing a comprehensive solution for deploying production-ready, self-managed MySQL infrastructure on Google Cloud Platform using OpenTofu/Terraform.

This isn’t just another “hello world” Terraform tutorial. We’re building enterprise-grade infrastructure with security-first principles, automated backups, and operational excellence baked in from day one.

• Blog URL : https://dcgmechanics.medium.com/building-production-ready-mysql-infrastructure-on-gcp-with-opentofu-terraform-a-complete-guide-912ee9fee0f8

• GitHub Repository : https://github.com/dcgmechanics/OPENTOFU-GCP-MYSQL-SELF-MANAGED

Please let me know if you find this blog and IaaC code helpful, any feedback is appreciated!

Thanks!

https://redd.it/1l2le99
@r_devops

Hep With Automatically Updating Database and Notification System

Hello. I'm slowly learning to code. I need help understanding the best way to structure and develop this project.

I would like to use exclusively python because its the only language I'm confident in. Is that okay?

My goal:

* I want to maintain a cloud-hosted database that updates automatically on a set schedule (hourly or semi hourly). I’m able to pull the data manually, but I’m struggling with setting up the automation and notification system.
* I want to run scripts when the database updates that monitor the database for certain conditions and send Telegram notifications when those conditions are met. So I can see it on my phone.
* This project is not data heavy and not resource intensive. It's not a bunch of data and its not complex triggers.

I've been using chatgpt as a resource to learn. Not code for me but I don't have enough knowledge to properly guide it on this and It's been guiding me in circles.

It has recommended me Railway as a cheap way to build this, but I'm having trouble implementing it. Is Railway even the best thing to use for my project or should I start over with something else?

In Railway I have my database setup and I don't have any problem writing the scripts. But I'm having trouble implementing an existing script to run every hour, I don't understand what service I need to create.

Any guidance is appreciated.

https://redd.it/1l2u9tz
@r_devops

Az400 Dumps

Anyone have Az-400 dumps???please share it with me my exam is tomorrow

https://redd.it/1l2u14b
@r_devops

DevOps Freelancer ? Let's connect

Hello Everyone,I am working as a Devops Engineer
in a start-up and it's completely remote. I get some time to upskill myself. I have close to one year of experience and I am planning to target FAANG after an year.
Currently I am looking for a side project or freelancing work . If you are interested in side project or doing some freelancing work already then I would love to understand the work and see if I can contribute

Also,If anyone can guide or suggest me something regarding the same , they are also free to DM.


Thank you !

https://redd.it/1l2wjvy
@r_devops

What are things that can scan for issues with your Dockerfile?

What are things that can scan for issues with your Dockerfile? Issues like outdated container, security flaws, etc.

https://redd.it/1l2wldk
@r_devops

Reducing IaC by 10x Without DSLs or Cloud Lock-In

One recurring challenge in multi-cloud infrastructure management is the amount of IaC required to support different cloud providers. Even minor architectural differences often demand full rewrites.

A more scalable approach is to define infrastructure using **provider-agnostic blueprints**, exposed through an SDK (currently Java, with C# in development). These blueprints encapsulate components, relationships, and governance rules as reusable, versioned modules.

In this model:

* Developers typically write \~3 lines of code:
* One to select the blueprint
* Two to connect their application services (e.g. a database, web app, or queue)
* No changes are needed when switching cloud providers
* Provider-specific details (region, naming, networking, identity) are handled inside the blueprint

Here’s an example of how an AKS cluster can be instantiated from a blueprint using a strongly typed SDK:

public static AzureKubernetesService getAks(String id, AzureResourceGroup resourceGroup) {

return AzureKubernetesService.builder()

.withId(id)

.withRegion(resourceGroup.getRegion())

.withResourceGroup(resourceGroup)

.withNodePools(getNodePools())

.build();

}

This approach shifts from a **“Live System First”** mindset (writing infrastructure tied to a specific cloud) to a **“Blueprint First”** model, where infrastructure is defined once and instantiated anywhere.

For example, a secure AKS cluster with observability and hub-spoke networking can be deployed in around 140 lines—with governance, compliance, and DR included.

The 10x IaC reduction isn’t about shortening code arbitrarily. It comes from **removing redundancy**, **centralizing expertise**, and **eliminating cloud-specific rewrites** for each use case.

**Anyone using provider-agnostic blueprints in production?**
How do you currently manage cloud divergence in your IaC workflows—custom modules, wrappers, or internal abstractions?

https://redd.it/1l31ejj
@r_devops

A tool for recognizing when getting close to limit for all aws resources?

Hey everyone.

My company uses many aws services. how can I know we're close to going over the limits? Building a function for each service is not sustainable, we need something dynamic. i can't just check the services we use, because sometimes developers will use a new service, and then adding that retroactively is not sustainable. any ideas?


edit- it's not about money, it's about sometimes there are hard limits of say 10 api calls per second, sometimes it's a soft limit that can be increased. how to keep up with this, when these limits are approaching?

https://redd.it/1l31a62
@r_devops

Cloud taught me to stop thinking like a “Python dev” and start thinking like a systems person

When I started doing cloud automation with Python, I approached everything like a typical dev:

Write a script

Handle exceptions

Make it reusable

Done ✅

But cloud work rewired me.

Suddenly i had to think about things i never used to worry about:

\>What happens if this Lambda retries?

\>Is this region even available right now?

\>Am I leaking infra costs through a loop i forgot to kill?

I had to zoom out.....past the code....and think like a systems person.
Python was still the tool, but the mindset had to evolve.

It was uncomfortable at first, but honestly?
It made me a way better engineer.

Anyone else feel this shift?


https://redd.it/1l32fp7
@r_devops

Does anyone in the DevOps world uses Bash?

Hey all,

Just wondering - being a DevOps myself for 10 years (and using Bash daily), is anyone still using Bash that heavily in todays world?

https://redd.it/1l3465f
@r_devops

Should I talk to my manager about my interest in DevOps?

I've recently started learning more about devops and it's implementation, I want to switch to a devops role eventually and at our current startup there is no dedicated devops engineer, we all just deploy manually and because of this I have a good understanding of deployment and its errors, there is no proper CI CD pipeline or containerisation and so on, I'm a software engineer with 2 YOE working on spring boot application mainly at present. Now I know it's not realistic to switch I just want to ask for more responsibility in that regard so I can learn and implement and also build my career. Is this ok? Am I rushing things? I've only started learning since 2 days

https://redd.it/1l364cz
@r_devops

How to set up Bitnami PostgreSQL-HA for multi-cluster replication with one primary and others as replicas?

I'm trying to build a multi-cluster PostgreSQL HA setup using the Bitnami postgresql-ha Helm chart.

Objective:

Primary cluster runs full HA (read/write)

Secondary clusters act as read-only replicas and should automatically follow the primary

If the primary region fails, a secondary should be promotable (manually or automated)

No manual replication config like modifying pghba.conf, primaryconninfo, or mounting standby.signal

Constraints:

Helm-based setup only

Cross-cluster replication must work out of the box or with Helm values

Has anyone successfully implemented this kind of architecture using Bitnami's charts or other Kubernetes-native PostgreSQL HA stacks (e.g., Stolon, CloudNativePG, Crunchy)?

Would love any pointers, Helm examples, or architectural suggestions that avoid drifting into manual setup territory.

https://redd.it/1l32p4w
@r_devops

AWS vs Azure Which Offers More Career Opportunities

I’m trying to decide which cloud provider to focus on. In terms of job market demand, growth potential, and career opportunities, which one offers more, AWS or Azure?

https://redd.it/1l39rgn
@r_devops

How do you divide responsibility between devs and ops for cluster instances vs app instances?

For companies that are striving for developer self-service where devs manage the app concerns and ops manage the lower level infra concerns, I have the following question:

How do you think about dividing responsibility between developers and ops for cluster instances vs app instances?

To me, it makes sense that developer should manage application cpu/memory and min/max instance count. But the cluster must be able to support that with sufficient instance sizes and count. So do you have the developers manage that too? Or do ops manage that, setting an upper bound on the limit. And to go beyond that, developers have to collaborate with ops to get that increased? Or something else like automatically set cluster max based on all the application max instance count?

https://redd.it/1l3aqsa
@r_devops

Need Help with DevOps Resume & Job Search


Hi all,
I’m a backend developer (2.5 years, C/C++, Linux) moving into DevOps. I’ve done some personal projects and got an AWS cert

Now I need help with:

What to put in experience section as I don't have devops exp in my current organisation

Making my resume DevOps-friendly

How to apply without real DevOps work experience

What kind of roles to target first


Any tips would be really helpful. Thanks!

https://redd.it/1l3cocn
@r_devops

Is DevOpsDays as a Noob worth it?

Hi, I saw there is a DevOpsDays event in my city coming soon, and recently the company I’m working at which is a startup offered me to be the DevOps for the team which I’m pretty excited about. However I don’t have that much experience, just a bit with AWS, I’ve been a developer for 2 years now. I was wondering if I ended up going to this DevOpsDays would I be lost during all the conferences or do you think I would be able to learn from them? I’ve never been to a conference before so I don’t know what they are like. Any recommendations?

https://redd.it/1l3dhlz
@r_devops

What non-technical DevOps / DX practices do you value most in your team?

Hello everyone,

after jumping from a ~5 person dev team to a ~100 person dev team recently and experiencing a different kind of team dynamic, I’ve been thinking a lot about the soft side of DevOps and DX, beyond just tooling and automation.

What are the softer and non-technical practices that your team adopted that made you happy as a dev? For example:

- how do you share business contexts and best practices
- how do you handle docs
- how do you get new devs up to speed and support them
- do you foster an engineering culture that pursues quality
- do you have someone you can always turn to for help

Curious to hear your good or bad experiences!




https://redd.it/1l3c9zt
@r_devops

Nuclei templates with AI

I would like to know about the increasing popularity of certain tools within the security domain, particularly in light of these agentic AI code editors and coding assistant LLMs. So, as of now my focus is on the use of Nuclei templates to automate the detection of vulnerabilities in web applications and APIs. How effectively can agentic AI or LLMs assist in writing Nuclei templates and has anyone successfully used these tools for this purpose?

So, i have a swagger specification and a postman collection of APIs although I know how to write Nuclei templates but I'm more curious if any LLMs or AI-based code editors could help me in this process. I understand that human intervention would still be necessary but even generating a base structure let's say, a template for detecting SQL injection would allow me to modify the payloads sent to the web application or specific API endpoints.

I would appreciate any insights from those currently using agentic AI code editors or LLMs to write nuclei templates and what the best practices are for leveraging such AIs in this context specifically

https://redd.it/1l3cfn5
@r_devops

Always the same?

We run our applications on openshift and as a devops guy I write the kubernetes deployment for applications and I do all the ops stuff. The deployment code is always the same: A bunch of deployments, secrets, cm, services etc. you need to template and a bunch of bash and python scripts chained together. Incidents are the same: „let’s write some simple queries in splunk or Prometheus to find the issue and then either write a simple fix like changing a config value we just googled or add a Prometheus alarm“
Every application feels same. It really doesn’t matter if it’s some data intensive application, an online shop or whatever.
I feel like no matter which technology I picked I only scratched the surface but can solve anything and there is no need to go deeper.

Am I the only one that feel so?





https://redd.it/1l3iyf1
@r_devops