Which Devops or cloud bootcamp or mentor to choose?

Hi everyone, I have some experience as a linux support engineer, product support technician and a bit of DevOps engineer, about 3 and a half years in total. I'm currently unemployed and want to get some real knowledge in practical terms to build and showcase some real projects. So far I bought myself KodeKloud pro subscription but it's not like a personal 1 on 1 plan where someone tracks and corrects me while doing stuff and that's what I'm missing.

I saw some reviews that people enrolled with Soleyman Shahir and their landing cloud roles, does anyone have any experience with his bootcamp?

I also saw Techworld with Nana, but from what i understood she doesn't have practical projects that build your portfolio and it kinda looks like more expensive version of KodeKloud to me...

Any recommendations or mentors please?

Best regards




https://redd.it/1lfjl12
@r_devops

Looking for recommendations on AWS SES + pinpoint

Hi Everyone. 

I'm an SRE working for a Medical Company. I have a question regarding SES + Pinpoint and its alternatives. I am working on a task for Federation, where I've been asked to track and show dashboard metrics to see the details of how many emails were opened / clicked/ rejected / complained / bounced / delivered. The requirement is to show how many are done, say in one month, and also which mail subject & email address it's been rejected. 

The current architecture is on keycloak - AWS SES - SNS - Cloudwatch - Datadog. It tracks and sends metrics on SNS and Cloudwatch. All the setup is done via terraform templates. I can see the open/click/etc details on both cloudwatch and datadog, but it's generic and doesn't include the specific details. 

I am tired of giving it via pinpoint, but since it's depreciated, my tf module rejects pinpoint_destination and the plan is failing. I tried creating a dashboard on datadog based on the query, but it cannot be restricted to an email address / subject. 

ChatGPT suggested that we use AWS Kinesis + firehose and show the dashboard based on the data stored in S3. The official documentation for Point recommends using Amazon Connect. While I'm working on that already, I'd like to know if there's a better way and if any of you are using such solutions already. 

Please share your thoughts. Have a wonderful day.

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

Sharing a guide on choosing cloud providers after seeing too many teams get stuck in analysis paralysis

Been working in the data space for a while and noticed a pattern... teams spend weeks comparing AWS vs Azure vs GCP feature lists like they're shopping for groceries, then still can't make a decision. It's frustrating to watch because the "perfect" comparison spreadsheet approach misses the actual point.

The reality is that the choice often comes down to strategic fit rather than who has the most services listed on their website. Take Netflix and Spotify as examples: Netflix runs on AWS while Spotify (similar scale/complexity) thrives on GCP.

My colleague put together a practical framework that cuts through the marketing noise and focuses on three key questions that actually matter:

1. What's your primary use case? (Not what looks cool, but what you need to ship)
2. How much infrastructure do you want to manage? (Some teams love control, others want to deploy and forget)
3. What does your team already know? (Retraining costs are real and underestimated)

The guide also includes a 30-day hands-on testing roadmap using free tiers, real cost gotchas to avoid, and examples of when each provider actually makes sense. Check it out here if you're dealing with this decision.

What's been your experience? Do you go all-in on one provider or mix them strategically? And has anyone here actually regretted their choice enough to migrate everything again?

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

CNAPP vendor got acquired, need alternatives - what's working for you?

Our CNAPP vendor just got acquired and we're already seeing problems. Alert volume has tripled with the same configurations, integrations are getting deprecated, and the product roadmap is now uncertain.

We're running mostly AWS with some GCP and Azure mixed in. The security team can't get a clear view across all our environments and we're drowning in alerts. Most of the high severity alerts used to be actionable, now we're spending too much time sorting through noise.

Need something that works across multiple clouds without locking us into one vendor. Must have solid API protection that can discover our endpoints automatically, and vulnerability management that helps us prioritize what actually matters. Runtime threat detection needs to work consistently whether we're on AWS, GCP, or Azure.

Has anyone migrated off a major CNAPP recently? What did you end up using and how's it working day-to-day? We're a team of 8 so the learning curve matters. Just want something that reduces alerts instead of creating more work.

Looking for actual user experiences, not sales pitches.

https://redd.it/1lfq2ie
@r_devops

We built an AI voice agent for DevOps as a joke.

First of all - I'll preface the entire post with this. You probably shouldn't use this. Not now, at least. Trusting non deterministic LLM's with your cloud account is the worst possible thing you could do.

We have tried ourselves and have also asked our friends/users, and the consensus is that the tooling just isn't ready to have folks prompt stuff into prod. Especially without an intermediary like terraform or pulumi, with versioning and what have you.

But about this voice agent thing, this whole thing started as a joke actually.

We were exploring Elevenlabs (no affiliation) and checking out how their voice API works. We had also been playing around with the AWS MCP server by Rafal Wilinski (also no affliation) for a while, so we thought, what would happen if we built a voice agent that could help us with AWS related stuff? (again, fully out of curiosity, and mostly as a joke)

This was the result: https://youtube.com/shorts/6PpBtWiEqiM?feature=share

Now, should this be used by folks? Probably not, lol.

But will voice agents be used in DevOps teams in the future? Maybe.

Most likely not for writing stuff onto your cloud account but for incident lifecycle management, runbook summarisation, new hire onboarding, cost summaries for execs, vulnerability checks, first line of support for devops teams, etc.

https://redd.it/1lfud92
@r_devops

Struggling to land a quality DevOps job.

I work in Kansas City and it's been very difficult to land a quality DevOps job. The latest job I've had for 8 months is abandoning IaC after a few years of trying. This job is now a career dead end and I'm struggling to find another growth position. This is the second job I've had where I specifically targeted it to learn these skills and it backfired. I came from a sysadmin background doing a mix of Linux and Windows systems enginering. Most jobs are either regular software engineering jobs writing product code or lousy systems admin jobs.

I manage some K8s clusters and write terraform code, but it's not a significant portion of our infrastructure.

Are DevOps jobs fairly rare outside of California?

https://redd.it/1lfv5qk
@r_devops

Is my CV (resume) bad, or is the job market just that bad right now in the UK?

I've been unemployed and job hunting for the last 4 months, and I've only managed to get 5 interviews. I'm going to run out of money fairly shortly and honestly I'm barely coping mentally.

I try to tailor my CV for any role that I find interesting, and for other roles I use this generic version of my CV: https://drive.proton.me/urls/EFEGBV146R#0SRZFnncaNIC

I've gotten exactly 0 interest from the above CV. My tailored ones look fairly similar, but I'll dive into more specific points/points I don't mention in the generic one above,. Feel free to destroy it.

If I don't get ghosted then I pretty quickly receive the "unfortunately" email we all know and love. 4 of my interviews didn't get past the first stage (always citing that there's a better candidate), and my 5th interview I did completely pass, but was rejected at the very end in favor of another person who passed... and that was for a type and size company I'm fairly certain I won't have another shot at for a very long time.

I feel I have a strong, diverse skill set, but I lack the knowledge and experience that comes from working at a higher-scale than I've been exposed to so far - I can't seem to find any company that would even consider taking a chance on me due to this. It makes me feel worthless.

Any criticism is appreciated, even the non-constructive kind.

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

Hey guys have been working on my opensource project, Guardian Platform - automated service discovery + multi-AWS account resource tracking

I have been facing this problem in my current work, where we have multiple repos, monorepos, all connected to each other but its hard for a new developer to understand what is what, how is it connected. I wanted a simple solution for this without overcomplicating so started on this project ->
https://github.com/sarim2000/guardian-platform

Also am trying to include cloud resources discovery in one place too (currently aws), since it was kinda hard for me to keep track of aws services and if multiple people are managing then then it does become a problem.

Will really appreciate feedbacks and what you think.

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

Live Stream - Argo CD 3.0 - Unlocking GitOps Excellence: Argo CD 3.0 and the Future of Promotions

Register Here:
Linkedin - https://www.linkedin.com/events/7333809748040925185/comments/
YouTube - https://www.youtube.com/watch?v=iE6q\_LHOIOQ

Katie Lamkin-Fulsher: Product Manager of Platform and Open Source @ Intuit Michael Crenshaw: Staff Software Developer @ Intuit and Lead Argo Project CD MaintainerArgo CD continues to evolve dramatically, and version 3.0 marks a significant milestone, bringing powerful enhancements to GitOps workflows. With increased security, improved best practices, optimized default settings, and streamlined release processes, Argo CD 3.0 makes managing complex deployments smoother, safer, and more reliable than ever.But we're not stopping there. The next frontier we're conquering is environment promotions—one of the most critical aspects of modern software delivery. Introducing GitOps Promoter from Argo Labs, a game-changing approach that simplifies complicated promotion processes, accelerates the usage of quality gates, and provides unmatched clarity into the deployment process. In this session, we'll explore the exciting advancements in Argo CD 3.0 and explore the possibilities of Argo Promotions. Whether you're looking to accelerate your team's velocity, reduce deployment risks, or simply achieve greater efficiency and transparency in your CI/CD pipelines, this talk will equip you with actionable insights to take your software delivery to the next level.

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

Bad situation at the workplace

Hi everyone, I need a little tip on the situation I'm living right now. I've been working as a "DevOps engineer" for about 9 months now. I quoted DevOps because I initially started an internship where I was promised to write Terraform modules, didn't end up doing that. I got to work with GitLab CI/CD, Python and Bash scripting, Helm and Kubernetes deployments. They hired me after the internship, but now I'm kind of in doubt on what to do. My team is basically just backend and frontend engineers, no one knows anything about DevOps except two guys in the backend that mentored me, but that's not their main thing. I got hired because the true Cloud Team of our company is extremely inefficient and apparently was never there when needed. Theoretically, I'm a backend engineer. In the meantime, I expanded myself (often upon force too, because I wanted to learn but they never let me expand too much) onto Terraform, monitoring and alerting with Prometheus and Grafana, ArgoCD, and I got to assist other people in deploying new applications outside my team as well.

I'm kind of getting to a point where I'm tired. Workplace is chill, colleagues are too, but I often don't have tasks/I create and assign them to myself. They let me do whatever I want basically, micromanagement doesn't exist because they simply don't understand much of what I do. I also think:
- Working mostly in one team reduces my capability of adapting to different tech stacks and assisting in other processes
- I do not have much freedom as much as I'd like. We have had Kaniko to build docker images in our CI/CD pipelines for two weeks after it's been deprecated, I've often brought up replacing it to multiple colleagues but they said it's not my job to do so.
- I wonder how much time I have left until I get fired? Things are already pretty stable with the changes and optimizations I've made to our cluster + monitoring etc.


Is this common? I know I should have seen the red flags since the beginning, but it was and still is my first job in IT and money is better than nothing. What should I do? Is my experience too limited to work in another company? I get recruiters on LinkedIn texting me but I'm scared it's bad offers/I'm not just able to compete with other people due to how limited my experience is.

https://redd.it/1lfz5du
@r_devops

Passing in a Kubernetes secret into a Helm Chart

Hello folks,

I am here in desperation. I can't seem to figure out how I can pass a variable/secret into a helm chart.

The secret, for example is like this (already created in advance):

apiVersion: v1
kind: Secret
metadata:
name: some-secret
namespace: somenamespace
type: Opaque
stringData:
TOKEN: "1233xxxxxx"

Then, my the Helm Chart I want to inject them in. Note this is an umbrella Helm Chart which just had the official one as a dependency.

templates/datasource.yaml

apiVersion: grafana.integreatly.org/v1beta1
kind: GrafanaDatasource
metadata:
name: prometheus-datasource
namespace: somenamespace
spec:
instanceSelector: {}
allowCrossNamespaceImport: true
datasource:
access: proxy
database: prometheus
jsonData:
timeInterval: 1m
enableSecureSocksProxy: true
secureSocksProxyUsername : "xxxxxxxx" # I need this to come from a that TOKEN in the secret
name: prometheus-local
type: prometheus
url: someurl:9090

I have spent countless hours and am still nowhere near an answer. It shouldn't be so tough

Help will be much appreciated

https://redd.it/1lg19kj
@r_devops

Can you give me suggestions for CD in Gitflow?

Hi all
I'm trying to define the CD of a Gitflow branch strategy.
What I want to define is when do the different Environments (dev, QA, UAT and prod) deployments trigger.
So far I'm thinking
Merge of any kind and from any branch to /develop triggers CD to Development
Branch creation or Push to /release branch triggers to UAT
Merge from /release or /hotfix to /main triggers to Prod with manual approval
Does that make sense?

What about QA? Maybe /develop with tags? Or /release_QA?

https://redd.it/1lg4zfq
@r_devops

Alternatives to JFrog Artifactory

Hi

I am/was a proponent of jfrog artifactory for small to middle (50 people) companies i contracted for. To install the self-hosted version for the following reasons:

* As a cache for artifacts (docker, maven, rpm, others) to put less stress on the internet uplink/downlink and to enable them to be able to work even when the/their internet is down. Main culprit here naturally CI/CD and developers.
* To store all inhouse artifacts they are legally required to keep for X years. Makes it easy to know what to backup and store.
* To store all inhouse artefacts (docker, rpm, maven, custom) with less stricts storage demands. Just so everyone knows where to go look for stuff.

Unfortunately JFrog for some unknown reason decided they want to get rid of the self-hosted installation method and told everyone to just use the cloud-hosted version. They told the companies they will retire self-hosted artifactory in the next 2-3 years. And doubled the price this year for the self-hosted license.

So here is the question: What are the alternatives? The hosted/cloud version is not an option.

I know there is nexus. Are there other options?

*Requirements*

Should be able to support several repository formats. The minimum is:

* docker
* maven
* rpm
* npm

Ideally these are also supported:

* generic (tgz or zip)
* python (pypi)

But naturally the more the better.


https://redd.it/1lg66wp
@r_devops

Building a Simple PaaS to provision EC2 instances from AMI's

https://youtu.be/qRy1x3iEHgo

https://redd.it/1lg7rjp
@r_devops

OpenTelemetry and Client Application Authenticity

Hi everyone, so... we would like to collect telemetry data from our mobile and web applications. We're stuck on how to verify authenticity of the client hitting our public otel collector. With backend applications we could somewhat trust the perimeter security where the services are inside the internal network. Firebase App Check https://firebase.google.com/docs/app-check seems promising as we use it in all our applications, and we should be able to use it in the otel collector endpoint. I just wonder if any one of you have implemented such a pipeline

https://redd.it/1lg9dtn
@r_devops

Related jobs that travel more

I work remotely, which is nice because I don't have to commute, but I would like a bit more variety. What jobs are tangential to DevOps that travel more?

https://redd.it/1lg5xsc
@r_devops

5 Years in DevOps and I’m choosing between 2 certifications

Hey Everybody,
I've been in DevOps for five years now, and I'm looking at a new certification. Need something for better pay, more job options, and just general career growth. I'm stuck between Red Hat and Kubernetes certs.
For Red Hat, I'm thinking about the RHCSA or RHCE. I've used Linux a lot, and Red Hat is known for solid enterprise stuff. But with everything going cloud native, I'm not sure how much a Red Hat cert still helps with job prospects or money.
Then there's Kubernetes. Looking at the KCNA for a start, or maybe jumping to the CKAD or CKA. Kubernetes is huge right now, feels like you need to know it. Which one of those Kube certs gives the most benefit for what I'm looking for? CKA for managing, CKAD for building, it's a bit confusing.
Trying to figure out if it's better to go with the deep Linux knowledge from Red Hat or jump fully into Kubernetes, which seems like the future.
Anyone got experience with these? What did you pick? Did it actually help with your salary or getting good jobs? Any thoughts on which path is smarter for the long run in DevOps would be really appreciated.




https://redd.it/1lgg8fj
@r_devops

Would an AWS infrastructure visualizer and security alerts all visualised via an interactive graph for less than 7 dollars a scan be useful?

As title states, i have built an aws infrastructure interactive graph visualizer and security violations. It works by using a read only iam role and scans all your aws resources using the necessary metadata and infrastruture. Its also runs your run of the mill security misconfigurations rules but also multi hop and complicated threats. For example privilege escalation etc. Which is what you can get with WIZ and others but pay a fraction of the price with mine
.as low as 5 dollars one time scan. wouldnt have runtime detection but can do real time scanning based on the iam role .

Is this something ppl would want?

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

Getting into devops

Hey so currently in a backend engineer internship and I'm currently coding, testing with postman, building with Jenkins, using grafana for testing.

I am enjoying it but maybe eventually I want to be dev ops. Can anyone help me with a good path for learning? And maybe certificates? Was hearing about the kubernetes certs. So any help would be appreciated

https://redd.it/1lgn6a3
@r_devops

How do you handle technical skill gaps in a managed services team supporting multiple Azure clients?

Hi everyone,

I work in a managed services company that supports multiple clients’ Azure environments. Our team handles tickets, incidents, and complex challenges, but we’re noticing a gap in technical depth across the team.

I’ve started using automation (emails, Teams, Power Platform) to improve ticket awareness, but I’d love to hear from others:

🔹 How do you address skill gaps in a busy support team?
🔹 What processes or tools have helped you upskill your engineers while still meeting client SLAs?
🔹 Any tips on balancing automation, documentation, and training?
🔹 How do you build a knowledge base that actually works?

Any real-world advice, examples, or lessons learned would be super helpful. Thanks in advance!

https://redd.it/1lgopn0
@r_devops

How to automate daily KPI emails from AWS CloudWatch using Outlook?

I’m working on a task where I need to fetch daily metrics from AWS CloudWatch for a few deployed models and send an automated status email via Outlook.

The metrics include:

4xx / 5xx Errors

API Latency (max & avg)

CPU and Memory Utilization

Total number of hits


I’ve got a fixed email template for this, and I currently send it manually every day. I want to automate the entire process — from pulling the data from CloudWatch to sending it via Outlook using a specific format.

I'm planning to use Python for this, probably with boto3 for AWS and win32com.client for Outlook email. Has anyone done something similar? Any best practices, sample scripts, or gotchas I should know about?

Would really appreciate your insights or any suggestions of youtube channel?

https://redd.it/1lgpfr1
@r_devops