Generative AI Engineer at Nokia
LLM assistants, retrieval workflows, and practical AI .
I build language-model systems that remain useful after the demo: grounded in retrieval, measured with evaluation, and shaped for real workflows.
- LLM assistants
- retrieval quality
- evaluation loops
- internal AI tools
01 / selected work
One project, properly explained.
graduation project · 2024
Super-Resolution for Digital Elevation Model Data
A comparative deep-learning study of SRGAN and ESRGAN for recovering terrain detail from low-resolution elevation data, including cross-region experiments between Türkiye and Austria.
- domain
- geospatial super-resolution
- models
- SRGAN + ESRGAN
- data
- SRTM · Türkiye / Austria
- evaluation
- PSNR · SSIM · MSE
02 / notes
Ideas worth working through.
From Contrastive Learning to World Models
A technical roadmap through visual self-supervised learning, from negative-pair contrastive objectives to explicit embedding regularization, latent prediction, and V-JEPA-style world models.
Notes on Reliable RAG Systems
A short placeholder note for future writing about retrieval quality, latency, and LLM assistants.
Super-Resolution for Digital Elevation Model Data
A graduation project on enhancing DEM resolution with SRGAN and ESRGAN, using terrain data from Turkiye and Austria.