Nursa AI-native transformation

Nenad took the risk of moving Nursa toward an AI-native way of building products, then led and executed the transformation across product strategy, proof, company adoption, and the operating model around delivery. The initiative was never limited to generating code faster. It changed how ideas were tested, how teams collaborated, how software reached production, and where human judgment stayed essential.

The work was cross-functional by design. Nenad led the broader product and organizational transformation in partnership with Nursa's AI technology and engineering leaders, who owned the technical architecture, implementation, and review practices behind the core-platform rebuild.

The transformation became credible when Nursa Study, a product for nursing students and colleges, went from idea to a working product in 48 hours. It was then developed into an enterprise offering with paying customers. Lovable documented that first proof and the wider shift it triggered inside Nursa.

The point was not the weekend build by itself. It showed that experienced product judgment, clear context, and AI-assisted execution could compress the path from problem to usable software. That proof gave Nursa a concrete reason to change how the company built, rather than relying on an abstract mandate to adopt AI.

The same initiative expanded from a focused product proof into Nursa's core healthcare staffing platform. In 2025, the team rebuilt the frontend, moved to a new technology stack, launched new product lines, and reconstructed a platform that had accumulated seven years of product history. The rebuild took seven months.

Lovable's story establishes the scale and speed of the wider transformation. CodeRabbit opens its case study by naming five leaders of the rebuild, including Nursa VP of Product Nenad Ivanovic, then states that this leadership group treated AI adoption as an operating-model shift. Together, the sources show one continuous initiative: prove the approach, scale it into the core platform, then redesign the systems that made that speed safe.

Coding agents increased output faster than traditional human review could absorb it. As Nenad explains in the CodeRabbit source: “We spent seven years building Nursa's core platform. This team rebuilt the whole platform from scratch in seven months using AI. It started to really hurt when we couldn't manage all of these code reviews alone; we needed AI to review the AI.”

Nursa moved to trunk-based development, where every pull request merged to main deploys directly to production. Automated review gave every change contextual baseline coverage. Experienced engineers kept authority over architecture, business rules, sensitive areas, and the changes that required deeper review.

During the rebuild, CodeRabbit caught three sensitive-data risks before merge and production: a full event object headed for a log line, a candidate email address being written to logs, and sensitive information rendered inside a failure-alert email.

The code worked and the tests passed. The risk was that personal or potentially regulated information would move into systems with different access and retention controls. Over six months, Nursa engineers accepted 65.4% of CodeRabbit's comments. The system reduced time-zone bottlenecks while preserving designated human approval for consequential areas.

The initiative extended beyond the core engineering team. More than 200 employees were encouraged to build with AI, including people in functions such as finance and credentialing. Nursa also began replacing roughly ten legacy SaaS tools as teams learned to create purpose-built internal software.

This is the leadership outcome behind the headline speed. Nenad did not personally engineer every part of the platform rebuild. He took the product and organizational risk, established the proof, led the AI-native transformation, and helped create the conditions in which product, engineering, and operational teams could execute it responsibly.