TNS
VOXPOP
Will real-time data processing replace batch processing?
At Confluent's user conference, Kafka co-creator Jay Kreps argued that stream processing would eventually supplant traditional methods of batch processing altogether.
Absolutely: Businesses operate in real-time and are looking to move their IT systems to real-time capabilities.
0%
Eventually: Enterprises will adopt technology slowly, so batch processing will be around for several more years.
0%
No way: Stream processing is a niche, and there will always be cases where batch processing is the only option.
0%
AI / DevOps / Software Development

Power up Your DevOps Workflow with AI and ChatGPT

ChatGPT is like having your very own super-smart, always-available tech buddy. The trick is learning to use this power effectively.
Jun 27th, 2023 6:17am by
Featued image for: Power up Your DevOps Workflow with AI and ChatGPT

Hey there, tech enthusiasts!

My name is Kelby Enevold, and I juggle all things tech training at Mission Cloud. My journey in the tech space spans about 14 years, and even though I’m not a self-proclaimed AI whiz, I’ve got some insights to share about a tool that’s been super helpful in my work. We’re all aware of the seismic shifts AI is causing across various sectors. So, are we heading toward an AI-dominated future where our jobs are at stake? Quite the contrary. AI, especially a nifty tool like OpenAI’s ChatGPT, is here to supercharge our workflows, not replace them.

AI: Your Productivity Supercharger

Let’s get one thing straight: AI isn’t some evil robot out to grab your job. If anything, it’s here to give you superpowers. Don’t get me wrong, AI will absolutely have a rippling impact on the service and product industry as a whole, but there will be an incredible amount of positive aspects to the overall adoption. One of the coolest things about AI models like ChatGPT is their potential to boost your workflow into overdrive, turning you into a productivity master.

Say you’re working on a new microservice in a Kubernetes environment. You’ve got to write tons of boilerplate code, and that can take a fair bit of time. With ChatGPT, you can simply ask it to whip up the necessary YAML files. This not only saves you time but also reduces the risk of errors from all that manual copying and pasting. However, remember this: It’s like using Stack Overflow. You wouldn’t just copy-paste the code you found there and call it a day, right? So, use the boilerplate code that ChatGPT provides as a starting point, then refine, iterate and test. Build on the examples you’re provided, double-check the documentation and see if it makes sense. Don’t fall into the pitfall of constantly copying and pasting solutions in hopes that they work on the first try.

Then there’s the age-old headache of troubleshooting. You’ve got a tricky server issue — maybe a misconfigured Docker container — and you’re stuck slogging through Docker documentation and Stack Overflow posts for hours. You’re feeling stuck, frustrated and your coffee’s gone cold. That’s where ChatGPT can swoop in and save the day. Feed it the error logs, and it can suggest potential fixes. It’s like having a personal troubleshooting assistant who can speed up the debugging process and save you from a ton of potential downtime. With recently added support for plugins and web browsing, this feature is becoming more useful than ever.

AI: Your Personal Code Guardian

AI tools like ChatGPT and Amazon CodeWhisperer aren’t just productivity wizards, they’re also your personal code guardians, so to speak. Tools like these help you keep your code in line with the industry standards, making sure you’re not inadvertently straying from best practices.

Let’s take AWS cloud infrastructure, for example. When you’re building one of those, you need to make sure you’re following the AWS Well-Architected Framework. Instead of manually cross-checking everything, you can ask an AI tool to validate your Infrastructure as Code (IaC) templates against the five pillars of the framework (operational excellence, security, reliability, performance efficiency and cost optimization). It’s like having a second set of eyes to ensure that your solutions are architecturally sound. Plus, these tools can even provide you with recommendations on how to improve your templates further.

AI: Your On-Demand Mentor

One of the most underrated aspects of ChatGPT is its potential as a killer learning tool. Let’s say you’re a junior DevOps engineer, fresh on the job. You’ve been tasked with setting up a CI/CD pipeline using Jenkins, but you’re not quite familiar with the technology yet.

Instead of spending hours trawling through Jenkins documentation or watching endless tutorials, you could simply ask ChatGPT about Jenkins. It can teach you how to set up a basic pipeline, demystify the intricacies of Jenkinsfiles and even give you tips on best practices. It’s like having an always-on-call mentor at your fingertips, helping you level up your skills at an accelerated pace.

All right, we’ve chatted a lot about the cool stuff AI can do, but it’s not all positive. We’ve got to talk about some of the possible downsides.

The Not-So-Bright Side of Large Language Models

While AI models like Bard and GPT-4 are incredibly powerful and versatile, they’re not perfect. One thing to remember is that these models can sometimes provide biased or inaccurate information. They’re trained on tons of data, and if that data has any biases or inaccuracies, they can unwittingly pass them onto you.

Another thing to remember is that while GPT-4 is impressively intelligent, it doesn’t always fully grasp the context of your specific DevOps project. It’s crucial to provide it with clear details and explain your use case to ensure ChatGPT really gets what you’re trying to do. ChatGPT contains training data up until 2021, meaning if you’re using a new tool or library, ChatGPT might not be aware and could use examples with deprecated libraries or features. So, always remember to double-check the info these tools give you against other reliable sources.

Lastly, let’s talk about dependency. GPT-4 is a fantastic tool, but relying on it too much might stifle your own problem-solving skills and creativity. You don’t want to end up using it as a crutch. Engage your brainpower, make use of traditional learning resources like documentation, knowledge base articles, online courses and don’t forget the invaluable knowledge exchange with your peers. The more you use models like CodeWhisperer and GPT-4, the more you’ll understand how to use them in a productive manner. Remember things like the context and overall goal are very important, as well as calling out limitations.

Conclusion

In the end, AI and tools aren’t here to replace us, but to supercharge our work and learning. As a testament to that, I’ve extensively usedChatGPT in refining this blog post. It was an active participant in my writing process, much like a collaborator or an editor. Whenever I was stuck on how to articulate a concept or if I needed help restructuring a paragraph, I’d turn to ChatGPT. I would ask for ways to express my ideas more clearly or for suggestions on organizing my thoughts better.

Perhaps what I appreciate most is how ChatGPT encouraged me to think critically about my writing. Just like an editor would, it challenged me to consider different angles and forced me to be clearer in my explanations. It didn’t do the job for me; it gave me a fresh perspective and helped me do my job better.

Even though it’s still early days, AI can offer a level of help and insight that outclasses traditional search engines and documentation. ChatGPT is like having your very own super-smart, always-available tech buddy. The trick is learning to use this power effectively while staying aware of potential pitfalls. Remember to stay curious, vigilant and keep on learning, because that’s what we do best. Enjoy the new tool you have to accompany you along your learning journey.

Group Created with Sketch.
TNS owner Insight Partners is an investor in: Pragma, Docker.
THE NEW STACK UPDATE A newsletter digest of the week’s most important stories & analyses.