Md Mostafizur R.

I am an AI Training & LLM Evaluation Specialist with a strong background in RLHF, data annotation, transcription, and multilingual content quality review. My professional track record includes successful project delivery for global leaders such as Google and Amazon, where I helped refine model accuracy and linguistic performance. I specialize in guideline-based tasks to improve chatbot performance, model alignment, and dataset accuracy for large-scale AI projects. Leveraging my 10+ years of experience as a translator and author, I provide the high-level reasoning necessary to identify model hallucinations and ensure perfect policy compliance. Core Skills: • LLM Evaluation & RLHF Feedback: Ranking model responses and fine-tuning AI reasoning. • Data Annotation & Classification: High-accuracy labeling for complex datasets. • Search Quality / Relevance Rating: Improving retrieval systems through expert evaluation. • Safety, Policy & Content Moderation: Auditing outputs for bias, toxicity, and compliance. • Prompt Testing & Evaluation: Stress-testing models to ensure reliable outputs. • Sentiment, Intent & Topic Tagging: Providing deep analytical context to linguistic data. What I Bring: ✔ Elite Experience: Proven history with Tier-1 tech giants (Google & Amazon). ✔ Linguistic Authority: 10+ years in translation, subtitling, and media localization. ✔ High Accuracy: Extreme detail orientation and strict adherence to complex guidelines. ✔ Consistent Delivery: Reliable performance on long-term, KPI-based tasks. Languages: • English — CEFR C1 (IELTS Certified) • Bengali — Native Speaker If you need a reliable evaluator to support AI training, RLHF workflows, multilingual data labeling, or quality-focused evaluation—I am ready to assist with your project.