- 👋 Hello, Namaste, नमस्कारः, Hallo, Salam, I’m @anjani-dhrangadhariya
- I build AI systems that make it into production. Most recently at Roche pRED, where I took a GenAI document generation pipeline from concept to validated production in a regulated environment, and built agentic extraction pipelines over scientific literature.
- My work sits at the intersection of NLP, LLMs, and life sciences: agentic workflows, RAG, information extraction, weak supervision, and semantics and FAIR standards. PhD in Computer Science (Geneva) on NLP and deep learning for systematic review automation, with 12 first-author publications.
- Interested in AI for clinical and scientific work that survives contact with real constraints — evaluation, validation, and getting people to actually use it.
📫 anjani.k.dhrangadhariya@gmail.com
- ♡♡♡ I love knowing about similarities in different languages, especially how Hindi and its mother, Sanskrit, are linked to Persian and its mother, Avestan, by the threads of history. Avestan and Sanskrit have several cognates and are mutually intelligible, forming deep links between ancient India and Iran far before the Arab conquest. Check out the repos @persian-hindi and @sanskrit-avestan. I am also interested in learning more about history and understanding how words metamorphosize over time within the Indian subcontinent. ♡♡♡