Open to opportunities

Sezer Sagir

@sezersagir

Message

I improve generative AI quality through RLHF auditing, LLM evaluation, and guideline development.

Germany

Message

What I'm looking for

I'm looking to contribute to generative AI quality through RLHF, LLM evaluation, error analysis, and guideline development, while continuing to build practical AI tools and grow my technical skills in information systems and cloud deployment.

I've audited annotator outputs and developed evaluation guidelines for large-scale generative AI projects via RWS and VolgaPartners, Outlier and co. maintaining a ~98% QA acceptance rate across 3,000+ complex code and logic evaluations for GPT, Claude, Gemini, and other major LLMs. My targeted feedback and error analysis reduced annotator error rates by 15–25% and improved team consistency by ~20%.

I'm currently pursuing a B.Sc. in Information Systems at KIT and build AI-powered tools in Python, including a TikTok Clip Analyzer using Llama 3.3 and a real-estate valuation tool deployed on AWS with Terraform.

Experience

Work history, roles, and key accomplishments

Current

AI Quality Assurance & RLHF Specialist

Current

Freelance

Apr 2025 - Present (1 year 6 months)

Audited ~120 annotator outputs per week at senior classification level within large-scale RLHF pipelines, reducing annotator error rates by 15-25% and improving team consistency by ~20%. Maintained a personal QA acceptance rate of ~98% across 3,000+ complex evaluations.

RLHF Pipeline LLM Evaluation Bias And Error Analysis Data Annotation Quality Assurance