kitUP is the flagship product of NativeMinds, an Istanbul app studio previously known as Ouro Media. What began as a single product has grown into a studio running more than twenty consumer apps, with new ones still in development. We spoke with Mehmet Can Özçelik, Data Engineer at NativeMinds, about what the data stack looked like before Bruin and what changed after. When he joined almost two and a half years ago, he was the only person on the data team.

Challenges Faced Before Bruin

NativeMinds' data work centers on one hard problem: pulling marketing spend from many sources, pulling transaction data from Stripe, and merging them correctly into the custom reports and dashboards the business runs on. Getting a number wrong is not an option when it drives spend decisions.

Before Bruin, that meant stitching a stack together by hand. The team pulled data with Fivetran, wrote Google Cloud Run functions to reach the sources Fivetran did not cover, and leaned on BigQuery scheduled queries to merge everything. Keeping it all correct and moving in a single flow was the constant struggle.

The architecture also capped how fresh the data could be. Between Fivetran's data pricing and the limits of Cloud Run functions and scheduled queries, the team could only run pipelines once a day. Anything on an hourly basis caused problems. With one or two products, that was tolerable. As the studio's app count grew, it stopped scaling.

Decision Points for Choosing Bruin

Bruin did not land on the first try. Fivetran was still answering the team's questions and the studio had only one or two products, so it was easy to set aside.

The moment it clicked was hourly data.

“The first time I realized Bruin was different was when we moved from daily to hourly pipelines and started getting Slack alerts and automations every hour. Suddenly I could tell the executives what was happening hour by hour.“

There was skepticism first. With working pipelines and only a couple of products, Mehmet Can was not convinced it was necessary and expected a painful adoption full of errors. The Bruin CLI and how quickly the team adopted it changed his mind.

“At first I thought we didn't need it. We only had one or two products and our pipelines worked. Now we have almost twenty products, and I think we should have done it even earlier.“

Impact of Bruin

The biggest change is how fast the team can stand up and adapt pipelines. What used to be a chain of manual steps, setting up each source, wiring it into Google Cloud functions, pulling code from an AI tool and re-implementing it in Cloud Run, now happens inside the tools Mehmet Can already works in.

“Adding a new product used to be a lot of steps. Now I work in Bruin from VS Code and Cursor, and it is so easy to adapt that we can add a new product fast. It is three or four times faster.“

Cost moved in the same direction. Bruin's pricing is structured around assets, and with far more products and assets today than the team had on Fivetran, the comparison is stark.

“With Fivetran, and we already had far fewer products back then, our cost would probably be four or five times higher than Bruin is today.“

Bruin also became part of how the team builds its own internal BI tooling. Mehmet Can uses Bruin AI to add Python functions and daily or hourly jobs, and to pull in his own documents and ingestors so new sources integrate quickly, whether he is working in Cursor or directly in the pipelines. That has kept a small, growing data team focused on real work rather than plumbing.

“I still handle integrations and operational work, but in between I just give one prompt to the AI and it already knows what to do. We have our documents and ingestors, so it integrates into Bruin quickly.“

What surprised him most was how little went wrong. He had braced for errors during adoption and across the growing set of products, and instead sees almost none on the technical side, backed by support that answers fast.

“I expected a lot of errors in the first phase. Instead we get almost none on the technical side, and the customer service is probably the best I've had. I ask something and get an answer in ten minutes to an hour.“

Tangible Benefits

• Pipelines that are three to four times faster to build and adapt, so new apps can be onboarded quickly from VS Code and Cursor.
• A data stack that costs up to five times less than the previous Fivetran, Cloud Run, and scheduled query setup, even while running many more products.
• A move from daily to hourly pipelines, with hourly Slack alerts that let the team report to executives in near real time.
• One platform behind more than twenty apps, replacing a hand-stitched mix of Fivetran, Google Cloud Run functions, and BigQuery scheduled queries.
• A small data team kept focused on integrations and analysis instead of infrastructure upkeep, with Bruin AI handling the repetitive setup.

Conclusion

NativeMinds started with one data engineer and one product and now runs a studio of more than twenty apps on Bruin. What was once a brittle, daily, hand-assembled stack is now a faster, cheaper, hourly one, and the team's main regret is not adopting it sooner.

“I'm happy using Bruin, and I've already recommended it to former colleagues. If anyone asks me, I definitely suggest it.“