Key Skills

Software

Internal/Proprietary Tooling

Top Subject Matter

LLM evaluation in the insurance domain

Biomedical NER

LLM evaluation

Top Data Types

Document

Text

Top Task Types

Classification

Entity Ner Classification

Fine Tuning

Prompt Response Writing SFT

Relationship

Freelancer Overview

My studies particularly focused on conversational analysis, spoken language processing, statistical analysis of linguistic data, natural language processing, and machine learning techniques. Regarding the acquired skills, the exam projects covered techniques such as text preprocessing, feature extraction, semantic similarity, sentiment analysis, text categorization, entity recognition and extraction, and relation extraction. During the internship, I collaborated with a company on a linguistic data extraction project in the medical/life science field, creating annotated datasets and a golden standard dataset. The project's goal is to identify and extract various types of biomedical entities from text, creating a link to a concept in a controlled knowledge system. The project's semantic network was integrated with the knowledge collected in the Unified Medical Language System. The master's thesis was carried out in collaboration with another company group, whose project concerned the company's linguistic model specific to the insurance domain, based on LLaMA-2, and the exploration of an alternative theoretical approach to its optimization. I created the domain-specific dataset for fine-tuning, based on text preprocessing techniques, entity recognition and extraction, and relation extraction, using proprietary Python libraries connected to the company's semantic network. I conducted an analysis of the results using standard performance metrics as well as other metrics.

Labeling Experience

Education

Work History