This is the first in a series of blog posts about How to build with GenAI- From strategy to implementation. In this series, we will explore the following questions:
- Is GenAI the right strategy for your product roadmap?
- Should you build or buy your GenAI model?
- How do you navigate the complexity of data to deliver clear results?
- Should you take a Human-in-the-loop approach?
- How do you manage costs while developing with GenAI?
***
It feels like everyone is talking about Gen AI right now, doesn’t it? From product announcements to AI-enhanced workflows, it’s become the buzzword of the moment. But the truth is, starting to build with Gen AI can feel overwhelming. Where do you begin?
At Yotascale, we asked ourselves that very question when we started our own AI journey to build our GenAI FinOps Yota co-pilot. As Jeff Harris, our Director of Strategy and Operations, shared during our recent webinar: “This space is so new. When we started, there wasn’t really an expert to call or a roadmap to follow. We had to figure it out as we went.”
Here’s how we navigated the early stages of adopting Gen AI into our product strategy—and what you can learn from our experience.
Step 1: Start with Exploration and Experimentation
The first step in any Gen AI journey is simple: explore and experiment. At Yotascale, we formed a small team and gave them the freedom to try things out. From testing API models to experimenting with fine-tuning and even hosting our own models, this phase was all about learning what was possible.
This hands-on approach helped us uncover the tools and techniques that worked for our needs—and just as importantly, those that didn’t.
As Jim Meyer, our VP of Engineering, explains: “The landscape feels overwhelming at first. But once you start experimenting, you get clarity about which solutions align with your product and your team’s capabilities.”
Don’t worry about getting everything perfect upfront. Use this phase to build a foundation of knowledge and confidence.
Step 2: Define Your Goals and Priorities
Once you’ve explored the landscape, it’s time to focus on what you want to achieve. Without clear goals, it’s easy to get sidetracked by shiny features or trending AI use cases.
For Yotascale, we started with one guiding question: What’s the most valuable problem AI can solve for our users? For us, that meant improving the user experience—abstracting away the manual drudgery of creating complex custom reports and simplifying the user experience to a simple conversational interface that anyone up and down the management chain can use. Cloud cost management software is incredibly complex, and the knowledge to generate custom reports is arcane and siloed. By enabling data access for anyone through simple conversations, we realized we could enable our customers to build the cloud conscious culture they want in their companies.
Jeff Harris puts it this way: “We knew AI wasn’t going to solve everything. But by zeroing in on our priorities, we could focus our resources where they’d make the biggest impact.”
Having a clear understanding of your goals will help you make smarter, more strategic decisions as you move forward.
Step 3: Stay Flexible and Keep Learning
One thing we’ve learned is that the Gen AI journey doesn’t stop once you pick a direction. The technology is evolving rapidly, and the best way to keep up is to stay curious and open to change.
Jeff adds, “You don’t need to solve every problem or compete with the latest innovations right away. Start with what makes the most sense for your product and users, and let that guide your path.”
This mindset allows you to adapt as the landscape shifts and new opportunities emerge.
Looking Ahead
Getting started with Gen AI is about more than just technology; it’s about understanding your goals, experimenting with possibilities, and staying adaptable.
In our next post, we’ll dive deeper into one of the most important decisions you’ll face: whether to build, buy, or fine-tune your AI model.
