Luka V.

Key Skills

Software

Other

Don't disclose

Top Subject Matter

NLP text classification and sentiment analysis training data

HCI research on LLM usage behavior (analysis of interaction data)

RAG chatbot development and training data preparation via retrieval/prompting

Top Data Types

Text

Document

Computer Code Programming

Audio

Video

Top Task Types

Classification

Text Generation

Computer Programming/Coding

Prompt + Response Writing (SFT)

Data Collection

Evaluation/Rating

Object Detection

Question Answering

Freelancer Overview

AI Developer – Classification & NLP Pipelines (FH OÖ Coursework, 2025). Brings 2+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include Internal, Proprietary Tooling, and Other. Education includes Bachelor of Science, University of Applied Sciences Upper Austria (FH OÖ) (2028) and High School Diploma, Svetozar Markovic Grammar School (2025). AI-training focus includes data types such as Text and labeling workflows including Computer Programming, Coding, and Evaluation.

Labeling Experience

AI Model Curation & Annotation Framework

TextClassificationData Collection

I designed and implemented a robust annotation framework focused on ensuring high-quality training data for AI model development. My work included the creation of labeling protocols and the execution of thorough data annotation in a research environment. I optimized workflows for prompt engineering and NLP curation using large language models. • Established high-fidelity classification protocols for AI training datasets. • Collaborated on prompt engineering for LLM data labeling workflows. • Maintained rigorous QA for data integrity in classification tasks. • Conducted hands-on annotation and model evaluation in a team setting.

2025 - Present

1. Categorized Dataset Cleaning & Preprocessing (Academic)

Don't discloseTextClassification

This project involved the development of a comprehensive annotation framework designed to generate high-fidelity ground-truth for machine learning classification models. The scope included the manual categorization and ranking of diverse datasets, specifically focused on optimizing Natural Language Processing (NLP) workflows and Large Language Model (LLM) performance. I was responsible for tagging and labeling thousands of data points to ensure robust training sets. Quality was maintained through strict validation protocols, cross-referencing annotations for 100% accuracy, and adhering to rigorous data integrity standards.

2024 - Present

Sentiment Analysis Dataset Development (Personal Project)

OtherTextClassificationPrompt + Response Writing (SFT)

Led analysis of large-scale AI assistant usage behavior by translating complex datasets into structures suitable for modeling. Built end-to-end preprocessing, cleaning, and visualization pipelines that prepare data for analytical interpretation and downstream AI training/validation. Produced executive-level dashboards from quantitative inputs to support data-driven conclusions from labeled/processed behavioral data. • Preprocessing and cleaning pipelines for behavioral datasets (text/interaction-derived signals). • Visualization and interpretation to support modeling decisions. • Data preparation workflows enabling model evaluation and training use. • Domain examples included medical diagnostics and regional pricing data contexts.

2025 - 2026

Education

Work History