Top 1 Percent DevOps and Cloud Engineering Skills | Must Watch

Summary

Highlights

• Abhishek emphasizes the burgeoning skills that keep you in top-tier DevOps positions. 💡
• MLOps, a transformative skill, automates the machine learning lifecycle similar to DevOps. 🤖
• AI Ops augments IT operations with predictive capabilities, reducing unexpected hurdles. 🔮
• PHOPS turns cloud cost estimation into an art, optimizing for better infrastructure spending. 📉
• Integrate AI assistants in DevOps to revolutionize workflow efficiency. 🤯
• Secure software supply chain practices are a must for any CI/CD focused professional. 🔒
• eBPF enhances the Linux kernel, facilitating better observability and system insights. 🔍
• Vasom bridges the platform gap in AI workload deployments, reducing developer headaches. 🌉
• Stay updated with advanced Kubernetes management strategies for large AI workloads. 🧠

Key Takeaways

• MLOps is the future! By focusing on machine learning lifecycle automation, you can differentiate yourself in the DevOps space. 🤖
• AI Ops can save the day by predicting future issues using AI on IT operations, making you indispensable to organizations. 🔮
• PHOPS, related to cloud cost optimization, is a promising skill that can transform you into a cost-savvy engineer. 🤑
• AI-assisted DevOps brings AI into DevOps activities, ensuring you're ahead in efficiency and innovation. 🤯
• Mastering the secure software supply chain can strengthen your CI/CD practices, safeguarding your projects. 🛡️
• eBPF upgrades Linux kernel capabilities, aiding in advanced observability and troubleshooting, a clear step forward! 🚀
• Vasom simplifies AI workload deployment across platforms, streamlining build and deployment processes. 🛠️
• Kubernetes management, especially for AI workloads, is crucial for handling large-scale clusters efficiently. 🧠

Overview

Chapters

• 00:00 - 00:30: Introduction Welcome to the channel hosted by Abhishek. In this introduction, he sets the stage for discussing key skills necessary to stay ahead in the fields of DevOps and cloud. Abhishek promises that these skills can help professionals remain in the top tier of their field. The chapter hints at an engaging explanation of these skills and their application.
• 00:30 - 01:30: Impact of AI on Skills The chapter discusses the rapid advancement of AI and its impact on skills. It highlights that while AI cannot replace any job currently due to its developmental stage, it poses a potential threat to certain skills in the future.
• 01:30 - 02:30: MLOps The chapter discusses the evolution of DevOps and cloud technologies, emphasizing their constant advancement.
• 02:30 - 03:30: AIOps The chapter titled 'AIOps' introduces the concept of MLOps, which stands for machine learning operations. Similar to DevOps, which automates the software development life cycle, MLOps aims to automate the machine learning life cycle. The focus is primarily on streamlining the process for developing machine learning models in companies.
• 03:30 - 04:30: PHOPS (Cloud Cost Management) The chapter discusses the emerging field of Cloud Cost Management, specifically targeting DevOps and MLOps engineers. It suggests that DevOps engineers can transition into MLOps by incorporating DevOps practices into the machine learning lifecycle. It highlights MLOps as a future skill to focus on for career advancement. It also briefly introduces the concept of AI ops, which involves applying AI to IT operations.
• 04:30 - 05:30: AI Assisted DevOps In this chapter titled 'AI Assisted DevOps', the focus is on the utilization of AI in operations, also known as AIOps. The chapter explains how AIOps can be used to predict future anomalies or potential issues within an organization. It provides a basic example scenario involving an e-commerce company. By analyzing metrics, logs, traces, and historical order data, the application of this solution can effectively forecast potential problems, enhancing operational efficiency and preemptive action plans.
• 05:30 - 06:30: Secure Software Supply Chain The chapter discusses the topic of preparing for unexpected or forecasted high volumes of orders within an organization. It explains the importance of scaling and taking appropriate actions based on forecasted demand, providing an example related to operations management (ops).
• 06:30 - 07:30: eBPF PHOPS (Predictive Optimization System) is essentially an evaluation of cloud billing. It involves understanding and estimating current cloud expenses and aims to reduce them. It can be considered a modern approach to cloud cost optimization, ensuring efficient management and optimization of cloud-related expenditures.
• 07:30 - 08:30: Vasom for AI Workloads The chapter titled 'Vasom for AI Workloads' discusses the current practices regarding cloud costs, billing, and reporting, as well as the use of dashboards. It highlights the evolving role of DevOps engineers who can easily transition into PHOPS engineers. The chapter emphasizes the growing importance of AI-assisted operations, pointing out that acquiring these skills will significantly enhance one's career prospects and resume.
• 08:30 - 10:00: Kubernetes Management for AI Workloads In this chapter, the discussion revolves around integrating AI into DevOps, specifically tailored for those already working in DevOps roles. The speaker suggests incorporating AI assistants and agents to enhance daily operations and mentions a complete playlist available for learning 'AI DevOps' in detail. There are over eight videos in the series, covering various topics from prompt engineering to the use of agents, with resources linked in the description.
• 10:00 - 11:00: Conclusion and Future Content In the conclusion chapter, the focus is on developing AI agents and working with them, which is a part of a larger playlist. It emphasizes the importance of DevOps as a crucial skill, alongside the secure software supply chain, which is an advanced version of tempecops.

Top 1 Percent DevOps and Cloud Engineering Skills | Must Watch Transcription

• 00:00 - 00:30 Hello everyone, my name is Abhishek and welcome back to my channel. In today's video, I'll talk about few skills that can keep you updated in the space of DevOps and cloud. Technically, these skills can keep you in the top 1% DevOps and cloud instance. This video is going to be interesting. I'll discuss what are these skills and I'll explain you how you can use these skills. So just make sure you
• 00:30 - 01:00 watch this video till the end. First and foremost, we all know how quickly AI is advancing right from assistants, agents to MCP. Recently, there is a lot of advancement and this is definitely threat to few skills in the future. I've told this multiple times. AI cannot replace any job at this point of time. you know it hasn't reached to that potential however in future it can
• 01:00 - 01:30 be threat to few domains but the important thing is that DevOps and cloud is constantly evolving what do I mean by that there are new verticals in the space of DevOps and cloud and if you learn these skills or if you learn these verticals you can keep yourself secure from AI or your job is secure from AI so let's quickly understand what are these skills. First one is
• 01:30 - 02:00 MLOps. What is MLOps? It is machine learning operations. So basically a lot of companies are focusing on developing their models. So as part of MLOps, you can automate the machine learning life cycle just like traditional DevOps which is focused on software development life cycle. MLOps focus on machine learning life cycle and
• 02:00 - 02:30 this is definitely a new opportunity. If you are a DevOps engineer, you already know what are the practices. If you incorporate these practices to machine learning life cycle, you can transition yourself into MLOps engineer or as a DevOps engineer, you can take that responsibility as well. So MLOps is one of the future skills that you can focus. Second one is AI ops. AO ops is AI for IT
• 02:30 - 03:00 operations. Basically using AIO ops, you can predict future anomalies or you can predict any future issues for your organization. A basic example let's say you work for e-commerce company and you have the metrics logs and traces or you have historical data of the orders for the e-commerce application using a obsolution you can predict any
• 03:00 - 03:30 unexpected number of orders or maybe you can tell your organization that at a particular date or at a particular time you can expect huge amount of orders and your company can be prepared for that basically by scaling according to the requirement or maybe taking the required action. This is a very basic example of a ops. Third one is
• 03:30 - 04:00 PHOPS. What is PHOPS? PHOPS is basically evaluating your cloud billing. PHOPS is more or less related to cloud. So understanding your cloud billing currently, estimating how much your company is currently spending on cloud and making sure your cloud billing is reduced. Some of the other way it is related to cloud cost optimization. So you know phops is modern cloud cost optimization but also
• 04:00 - 04:30 deals with current cloud cost or the current billing and reporting as well as dashboards. This is a very promising skill if you ask me. So basically if you're a DevOps engineer, it's very easy to transition yourself into a PHOPS engineer in future or just learn that skill and it's going to add a great addition to your resume. Fourth one is AI assisted
• 04:30 - 05:00 DevOps. Now this is very very simple. If you're already a DevOps engineer, just incorporate AI assistants, AI agents into your day-to-day activities and even prompt engineering using which you can learn AI DevOps. I've done a complete playlist on it. There are already eight videos in the playlist or even more than that. I'll put the link in the description. Right from prompt engineering to using agents using crew
• 05:00 - 05:30 AI developing your own agents working with the agents you can learn all of that in the playlist. So a devops is the fourth skill that you can focus on. Fifth one is secure supply chain or secure software supply chain. what we can learn as part of this skill you know it is advanced version of tempecops or it's a skill that is focused on
• 05:30 - 06:00 CI/CD if you're already working on CI/CD you can just make sure you track all the dependencies of your application and ensure the direct dependencies or transitive dependencies of your application are secure there are few things like sbomb software will of materials and there are few tools that can help you in this space. It's a very simple addition if you are already
• 06:00 - 06:30 working in the space of CI/CD. This is the fifth skill. Sixth one, EBPF. Now this is a very very interesting one. Of course it is little complicated but if you have some time you can start understanding what is EBPF. This particular skill focuses on advancing or enhancing your Linux kernel. Linux kernel development is not
• 06:30 - 07:00 that easy. You know very rarely you see uh improvements in the Linux kernel because it's very difficult to update it add new features to it. But ebpf is a method where you can indirectly add new features to your Linux kernel and this is going to help you in the space of observability. Every transaction goes through kernel. So it helps you in the distributed tracing metrics as well as
• 07:00 - 07:30 logs. In fact in future you might reach to a point where using ebpf you don't even have to define any metrics logs any metrics or traces in your application. You know this is just a prediction but at this point of time if you already have metrics logs and traces ebpf is going to provide more details about your application. So it's just like advanced observability.
• 07:30 - 08:00 So this is skill number six, Vasom. So Vasom is also something very promising. It helps you deploy your AI workloads onto the Kubernetes cluster or even Docker containers. So today even OpenAI deploys their workloads onto Kubernetes cluster. So you know building and deploying the machine learning workloads or AI workloads is not that easy. Looking at
• 08:00 - 08:30 different platforms that you have today whether it's x64 or whether it is ARM every time building so many AI workloads on different platforms is not that easy. Vasom acts as a layer and you don't have to build your AI workload for different platforms for a lot of platforms. So it is going to reduce your efforts in the space of build and deployment especially
• 08:30 - 09:00 related to AI workloads. So vom is one of the skill that you can focus on. Finally, Kubernetes management. Mhishek, we already work on Kubernetes cluster upgrades or Kubernetes management. But Kubernetes management with respective AI workloads. Let me make it simple. Today you already work on managing your Kubernetes cluster. But in future the
• 09:00 - 09:30 size of Kubernetes cluster is going to be huge because if you want to deploy machine learning workloads the size of cluster is going to be high and number of name spaces on the cluster is also going to be high. This is where you see advancement of vcluster or projects like cubsaw. So using your Kubernetes name spaces as tenants is something that you do today but using your Kubernetes
• 09:30 - 10:00 namespaces as virtual clusters is going to grow. So you can learn things like vcluster cube saw to keep yourself updated in the space of devops and cloud. Of course I'll make more videos on these topics because right now I'm just explaining them in the oneliners. But if you are really interested I'll make videos on individual topics and I'm also going to cover a lot of tools like for MLOps we will learn about
• 10:00 - 10:30 MLflow cubeflow for kubernetes management we will learn about vclustluster cube saw when it comes to phobs I'll talk more about open cost AI so there are different tools in this space in each of these skills There are multiple tools that are evolving. So I have plans to make more videos on them. So yeah, if you have any questions related to this topic, I'm more than
• 10:30 - 11:00 happy to help. Do let me know in the comment section. See you all in the next one. Take care. Bye-bye.