My name is Sina.
I have just finished my master's in Data Science and Computational Linguistics from UBC. I am currently doing an internship building stateful agentic workflows for Transcarent.
My speciality is NLP. I care about building safe, production-ready language models.
The three projects on my profile reflect this:
- The Linguistic Examination project is about identifying meaningful linguistics features to predict the native language of English learners.
- Gemini Guardrail is about assessing the accuracy of summaries produced by Gemini- ensuring reliability and safety.
- Membership Inference Attack (LLM & VLM) is about assessing the robustness and privacy of models.
Current Focus | Building scalable RAG (Retrieval-Augmented Generation) systems. Orchestrating stateful agentic workflows.
- 📫 How to reach me: sina.oloomi@mail.mcgill.ca