RigUp Innovation, Technology & Agility

What is the unique story that you feel your company has with AI? If you were writing about it, what would the title of your blog be?
I’d say our story is about starting with the layups and expanding methodically, both in the product and internally, by giving our team the right tools and support to bring real value to our clients. We looked hard at what we’d built and found clear spots where AI could replace or improve it, like automating document extraction tasks. If we were writing about it, I would say the title of the blog would be “Validating AI Use Cases Through Fast, Focused Prototyping.” What makes our story unique is the industry we’re in and the massive opportunity AI opens up here. Trust and accuracy around client/vendor transactions and time tracking in the energy industry have always been tough. AI lets us handle both structured and unstructured data faster, more accurately and at scale.

What was a monumental moment for your team when it comes to your work with AI?
A key moment for us was realizing how fast applied AI could improve tasks we’d been handling with custom, templated workflows. Early on, we focused on no-brainer use cases — areas where AI models were already strong. We ran quick tests and right away saw legitimate gains in speed and accuracy. It wasn’t plug-and-play — we still had to tailor it to our needs — but it proved the value quickly and guided us toward investing further in that space. That led to a shift in how we work: we now emphasize fast, focused experimentation. Build proof of concepts, validate them and either scale or move on. That mindset helps us spot the right problems to solve without getting stuck chasing AI just for the sake of it.

What challenges did your team overcome in AI adoption?
Our leadership has been really supportive in this space, both for our small AI team and our employees who are leveraging the tools we’ve built or deployed for them. We try to keep model and provider bias low and develop a code architecture that allows us to plug and play the latest models and releases from any provider into our current workflow and get results almost immediately after the models become available.

That’s one way we try to keep up, but I won’t sugarcoat the current state of AI. It’s moving at breakneck speed and it’s challenging to keep up with. It requires a lot of blog reading, experimentation, knowledge sharing and what feels like a constant check on your understanding and bias. For example, model context protocol servers are in the category of a new hot trend in AI, and as we’ve started experimenting, other solutions or tools that we have spent a lot of time and effort on are rendered out of date or not reasonable to use given the existence of new AI trends, features or capabilities.