Fjolla Berisha

Key Skills

Software

Telus

Mindrift

Top Subject Matter

Natural Language Processing

LLM Evaluation

Generative AI

Top Data Types

Text

Document

Audio

Top Task Types

Prompt Response Writing SFT

Classification

Object Detection

Text Generation

Question Answering

Computer Programming Coding

Freelancer Overview

I have direct experience in AI model training and evaluation from my roles at Alignerr/Telus International (Prompt Engineer, 2025–present) and Mindrift (AI Tutor, 2024), where I engineered prompts, annotated datasets, and evaluated LLM outputs for accuracy, coherence, and safety. I hold an MSc in Computer Engineering and am currently completing a PhD in Computer Science, with strong skills in Python, Java, SQL, data analysis, and machine learning frameworks (TensorFlow, Scikit-learn, PyTorch).

Labeling Experience

AI Prompt Engineer

TelusTextPrompt Response Writing SFT

As an AI Prompt Engineer at Telus International / Alignerr, I engineered prompts and evaluated large language model outputs for accuracy, coherence, and safety. I performed structured annotation and behavioral evaluation of LLMs as part of applied AI research workflows. The role focused on evaluating and improving generative model outputs for safety and quality. • Developed and refined prompt templates for LLM evaluation. • Conducted annotation and rating of LLM-generated text outputs. • Collaborated with researchers to iterate evaluation criteria. • Ensured outputs met safety and factuality standards.

As an AI Prompt Engineer at Telus International / Alignerr, I engineered prompts and evaluated large language model outputs for accuracy, coherence, and safety. I performed structured annotation and behavioral evaluation of LLMs as part of applied AI research workflows. The role focused on evaluating and improving generative model outputs for safety and quality. • Developed and refined prompt templates for LLM evaluation. • Conducted annotation and rating of LLM-generated text outputs. • Collaborated with researchers to iterate evaluation criteria. • Ensured outputs met safety and factuality standards.

2024 - 2025

AI Tutor

MindriftText

As an AI Tutor at Mindrift, I trained generative AI models and evaluated outputs for factual accuracy, coherence, and quality. The role entailed providing feedback on model-generated text and instructing learners in Python, data analytics, and applied AI concepts. Core activities included hands-on rating, annotation, and instructional duties relevant to AI training. • Evaluated and rated model-generated outputs for accuracy. • Provided instructional support for learners in AI-related subjects. • Utilized internal assessment tools for data annotation tasks. • Contributed feedback to improve AI model performance.

As an AI Tutor at Mindrift, I trained generative AI models and evaluated outputs for factual accuracy, coherence, and quality. The role entailed providing feedback on model-generated text and instructing learners in Python, data analytics, and applied AI concepts. Core activities included hands-on rating, annotation, and instructional duties relevant to AI training. • Evaluated and rated model-generated outputs for accuracy. • Provided instructional support for learners in AI-related subjects. • Utilized internal assessment tools for data annotation tasks. • Contributed feedback to improve AI model performance.

2024 - 2024

Education

Work History