Exploring the frontier of Natural Language Processing and Agentic Systems to bridge the gap between human intent and machine execution.
Welcome to my GitHub profile!
I am a \ an ...
-
Software & Data Engineering Working Student @Doc Cirrus GmbH -
MSc Data Science student @Berliner Hochschule für Technik (BHT). -
Deep Learning & NLP Engineer leveraging a strong Full-Stack experience to build scalable AI applications. -
Full-Stack Engineer with 2+ years of experience in the Ed-Tech domain. -
Open to collaborative projects that tackle real-world problems using Machine Learning and modern Software Engineering practices.
With a follow up paper to extend the research,
This research addresses the significant problems caused by handwritten drug prescriptions, such as medication errors from misinterpretation and the lack of standardized data for research.
Feel free to reach out, I'll be happy to hear from you.
The images shown above are cached snapshots stored in images/fallback/. A scheduled GitHub Action (.github/workflows/cache-badges.yml) attempts to fetch the live badge images daily and commits updated snapshots when available. If a remote service is unreachable (HTTP 503 or similar), the last successful snapshot remains displayed.
The workflow runs on push and every 6 hours by default. It uses conditional requests (ETag / Last-Modified) and stores per-file metadata alongside each cached image as images/fallback/<name>.etag and images/fallback/<name>.lm to avoid unnecessary downloads.
