Computational Linguist

Making multilingual AI and data more reliable

I evaluate LLM behavior, validate multilingual research data, and turn recurring language-quality issues into robust, transparent workflows.

Get in touch

What I do

How I can help

A focused look at where I create the most value.

LLM Evaluation

I assess model behavior and failure patterns in literary-text analysis, with a focus on capabilities and limitations.

Linguistic QA

I validate multilingual corpora and design pattern-based checks to surface systematic data-quality issues.

Research Workflows

I build digital research infrastructure, visualizations, and open-science knowledge resources.

About

Language expertise, applied

Evgenia Fileva — Computational Linguist

I am a computational linguist with five years of experience in multilingual NLP and research-data validation at the Trier Center for Digital Humanities. My work includes independent research on GPT-family models for literary-text analysis and recurring failure patterns.

My work combines linguistic precision, multilingual analysis, and practical tooling. I am currently completing full-time training in AI agents and workflow automation at WBS through September 2026.

Based in: Trier, Germany
Background: 5 years in multilingual NLP and research data
Languages: German (C1) · English (C1) · Russian (native)
Focus: LLM evaluation, AI agents and workflow automation

Previously at

Trier Center for Digital Humanities · CLS INFRA · University of Trier · National Research University, Samara · Samara State University

Contact

Let’s talk

Let’s connect about LLM evaluation, multilingual NLP, or AI workflow automation.

Get in touch

Trier, Germany