Full-Stack Developer
NLP Researcher & Designer
— Bonn, Germany
Hi, I'm MahshidI build intelligent thingsthat actually work
I'm one of the few people who can fine-tune a language model in the morning and ship a production React dashboard in the afternoon. That's not an accident — it's what I've been building toward.
Most engineers pick a side. I didn't. Two years of NLP research at Universität Bonn alongside production work at Fraunhofer IEM taught me that the best products get built by people who understand both the model and the interface it powers.
Currently pursuing an M.Sc. in Computer Science at Universität Bonn while working as a Software Developer at Fraunhofer IEM.
Where I've shipped
real things
Fraunhofer IEM
Software Developer (Werkstudentin)
- Migrated build system from CRA to Vite, cutting cold-start times by ~20× and improving CI/CD pipeline efficiency.
- Standardized UI across 20+ pages to MUI design system, establishing a reusable component library that reduced design-to-dev handoff friction.
- Refactored data tables to Material React Table with virtual scrolling, handling 1,000+ row datasets without pagination overhead.
- Built 5 interactive data visualization dashboards for engineering teams, enabling exploration of complex multi-dimensional datasets.
- Built and optimized Flask API endpoints and SQL queries to power data retrieval and aggregation across engineering dashboards.
- Designed and shipped a scalable Figma Design System, adopted as the UI foundation across the product.
Pishma Company
UI/UX Designer (Part-time)
- Created Figma interfaces and prototypes for 3+ digital products, improving consistency and iteration speed.
- Built a Design System adopted across 4 products and translated usability research into measurable UI improvements.
Where I've learned
to think better
Universität Bonn
M.Sc. in Computer Science (Informatik)
Research Projects
Studied cross-lingual bias in LLM judges across 14 languages and 33K score pairs. Fine-tuned LoRA adapters with Qwen2.5-14B, reducing Arabic scoring bias by 72%.
Fine-tuned RoBERTa and XLM-RoBERTa to classify 40+ narrative labels on multilingual datasets, achieving F1 up to 0.69 with robust cross-lingual transfer.
Azad University, Science and Research Branch
B.Sc. in Computer Engineering
Frontend Engineering
I don't just build UIs, I build the systems that make UIs maintainable. At Fraunhofer IEM, I migrated a legacy CRA codebase to Vite (20× faster builds), standardized 20+ pages to MUI, and shipped a component library that eliminated design-to-dev friction across the team.
Backend & APIs
My frontend work taught me exactly where backends break. At Fraunhofer IEM, I built Flask API endpoints and optimized SQL queries powering engineering dashboards used daily by the team. I write backend code because I need it to work, not because it's my primary identity.
ML & NLP Research
I work on the problems that don't have Stack Overflow answers. At the University of Bonn AI Alignment Lab, I reduced cross-lingual scoring bias by 72% across 14 languages by fine-tuning LoRA adapters on Qwen2.5-14B. Before that, multilingual narrative classifiers across 40+ labels with F1 up to 0.69. This is where I learned that rigour isn't optional.
Design Systems & UI/UX
Design systems are engineering problems. I've built Figma-to-code pipelines adopted across 4 products at Pishma, and a component architecture at Fraunhofer IEM that standardized UI across 20+ pages. The goal is always the same: make good design the path of least resistance for every developer on the team.
What I stand for
The principles I keep coming back to, the ones that shape how I think, build, and collaborate.
01
Systems over chaos
I build infrastructure meant to outlast me. Reusable components, typed interfaces, and design systems that new teammates can read in an afternoon. The structural work nobody notices is usually the work that saves thousands of hours.
02
Evidence beats intuition
I don't ship on gut feeling. Every decision, whether it's an NLP fine-tuning choice or a UI layout, is backed by data, testing, or research. Opinions are starting points. Results are the argument.
03
Rigour without jargon
Code that works isn't enough if nobody can use what it powers. Engineering precision and interface clarity aren't competing values. They're the same standard applied at different layers.
04
Research informs. Shipping teaches.
I work at the intersection of NLP research and production frontend, which means I'm always a student in one direction. Curiosity compounds. The fastest way to learn something is to ship it.