Mario A. K.

@mariokade

AI Data, LLM Evaluation QA Specialist

Griekenland
Grieks, Engels
Sommige informatie wordt in het Engels weergegeven.
Over mij
I am a multilingual AI Data, LLM Evaluation, and QA specialist with over 4 years of experience. I specialize in evaluating AI-generated content for factuality, safety, and instruction adherence, alongside technical expertise in Python, Linux, and cloud infrastructure. I have a strong background in technical support and compliance operations.... Lees meer

Skills

m
mariokade
Mario A. K.
offline • 
Gemiddelde reactietijd: 1 uur

Bekijk mijn diensten

Websites op maat
I will turn your educational content into an interactive web experience
VanafUS$ 80 / projectMeer informatie
Gegevenslabeling en -annotatie
I will evaluate your ai responses in greek for accuracy, naturalness and safety
VanafUS$ 15 / projectMeer informatie

Portfolio

Werkervaring

AI Preference Ranking Evaluator (RLHF)

SuperAnnotate

Apr 2026 - Present • 4 mos

• Perform cross-lingual pairwise ranking for a major LLM. • Evaluate instruction execution, JSON/Markdown formatting compliance and tone. • Provide structured judgments distinguishing quality across competing model responses.

Welocalize

Greek LLM Evaluator (RLHF)

Welocalize

May 2026 - Aug 2026 • 3 mos

• Conduct multi-stage pairwise preference evaluations of Greek-language AI responses. • Assess outputs for factuality, safety, linguistic naturalness and overall response quality. • Apply detailed evaluation criteria consistently across large AI-generated response sets.

Mindrift

LLM Output Evaluator & Prompt Engineer

Mindrift

Dec 2025 - Apr 2026 • 4 mos

• Designed prompts for model capability benchmarking and evaluated outputs against client specifications. • Assessed instruction adherence and supported AI training/evaluation workflows through structured feedback.