Dan is a software engineer skilled in building and operating distributed systems that process more than 1 billion messages per day, and in making AI systems reliable and trustworthy.

Before joining Janea Systems, he was at Infobip for 8 years, where he built machine learning systems for fraud detection and chatbots, all before large language models. More recently, he built an evaluation pipeline for AI chatbots grounded in retrieved data that provided actionable insights in minutes, instead of days when done manually.

Following the acquisition of a US company, Dan led a rebuild of its legacy Java system for managing SMS campaigns and executed the client migration. He also resolved queueing bottlenecks on the SMS platform, doubling its throughput to 60k messages per second. After these successes, he took on responsibility for leading a team that owned the shared AI infrastructure, enabling other teams to integrate LLMs into existing products and providing observability and billing layers.

For the current chapter of his career, Dan is looking to apply his experience to different domains and help teams build reliable systems that make a positive impact.

Dan holds a master’s degree in computer science from the University of Zagreb. When not working, he enjoys hiking, sailing, and learning about our history.