Professional
NDANLP Tools for Quality Management
At BMW, I designed and built an internal AI chat interface that later grew into a wider NLP toolkit for quality management.
I worked between UX and frontend development, turning complex AI results into interfaces people could understand and use in their daily quality management work.
The short version
- Took the first chat experience from Figma prototype to Angular implementation
- Around 100 people at BMW used or tested the growing product family
- Worked with more than 10 teams, including colleagues in Spartanburg
What I worked on
- I used interviews and workshops to understand existing workflows, then turned the findings into epics, user stories and clear acceptance criteria.
- I designed filters, supporting text and progressive disclosure, then built the Angular components in TypeScript and connected them to Java services.
What came out of it
What started as my internship chat prototype became a family of tools for several departments. I stayed with the product for 17 months and helped it grow as the team and use cases expanded.
What I learned
The project showed me that displaying an AI result is not enough. People also need to understand what it means, how reliable it is and where its limits are. Starting with their real workflow often led us to simpler features. I also adapted my documentation for teams with different working styles.
Tools and methods
- Angular
- TypeScript
- Java
- Figma
- Docker
- Confluence
- BitBucket
- Git