Looks like you have JavaScript disabled. For the full Twine experience, you will need to re-enable it.

AI Data Trainer and LLM Prompt Engineer specialising in large language model evaluation, prompt design, and RLHF-aligned data annotation. I build multi-turn conversational prompts, evaluate model outputs with rubric-based defect analysis, and deliver clear, defensible rationales for iterative improvements. I also bring strong operational QA and customer-quality experience—ensuring privacy/NDA compliance, high accuracy in high-volume workflows, and consistent quality auditing across Hindi and English transcription and conversational AI testing.…AI Data Trainer and LLM Prompt Engineer specialising in large language model evaluation, prompt design, and RLHF-aligned data annotation. I build multi-turn conversational prompts, evaluate model outputs with rubric-based defect analysis, and deliver clear, defensible rationales for iterative improvements. I also bring strong operational QA and customer-quality experience—ensuring privacy/NDA compliance, high accuracy in high-volume workflows, and consistent quality auditing across Hindi and English transcription and conversational AI testing.

DEEKSHA GUPTA

AI Prompt Engineer, AI Evaluator, AI Trainer, +1





AI Data Trainer and LLM Prompt Engineer specialising in large language model evaluation, prompt design, and RLHF-aligned data annotation. I build multi-turn conversational prompts, evaluate model outputs with rubric-based defect analysis, and deliver clear, defensible rationales for iterative improvements. I also bring strong operational QA and customer-quality experience—ensuring privacy/NDA compliance, high accuracy in high-volume workflows, and consistent quality auditing across Hindi and English transcription and conversational AI testing.…AI Data Trainer and LLM Prompt Engineer specialising in large language model evaluation, prompt design, and RLHF-aligned data annotation. I build multi-turn conversational prompts, evaluate model outputs with rubric-based defect analysis, and deliver clear, defensible rationales for iterative improvements. I also bring strong operational QA and customer-quality experience—ensuring privacy/NDA compliance, high accuracy in high-volume workflows, and consistent quality auditing across Hindi and English transcription and conversational AI testing.

Available to hire

AI Data Trainer and LLM Prompt Engineer specialising in large language model evaluation, prompt design, and RLHF-aligned data annotation. I build multi-turn conversational prompts, evaluate model outputs with rubric-based defect analysis, and deliver clear, defensible rationales for iterative improvements.

I also bring strong operational QA and customer-quality experience—ensuring privacy/NDA compliance, high accuracy in high-volume workflows, and consistent quality auditing across Hindi and English transcription and conversational AI testing.

Skills

Experience Level

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Intermediate

Language

Hindi

Fluent

English

Advanced

Work Experience

AI Data Trainer & LLM Prompt Engineer at Freelance / Contract Projects

January 1, 2023 - Present

Designed and executed multi-turn conversational prompts (3–8 turns) for LLM evaluation across domains, assessing contextual coherence, personalisation accuracy, and response helpfulness. Evaluated model outputs using structured rubrics for grounding, hallucination rate, tone accuracy, and instruction-following, providing detailed written rationales. Conducted side-by-side (SxS) response comparisons to stack-rank model outputs and documented recurring defect patterns with actionable improvement feedback. Identified and flagged hallucinations, factual errors, grounding failures, and inappropriate personalisation in Hindi and English outputs. Maintained strict data hygiene by purging sessions post-evaluation and following privacy and NDA guidelines. Contributed to RLHF pipelines via structured feedback and annotated/QA-reviewed datasets (prompt-response pairs, dialogue trees, preference rankings) achieving 97%+ inter-annotator agreement.

Team Leader — Operations & Quality at Flipkart

January 1, 2021 - December 31, 2023

Led quality-focused operations for a 15+ member team, implementing structured daily briefings and real-time performance dashboards. Authored SOPs, structured feedback reports, and reusable training modules adopted across multiple teams. Resolved escalations using root-cause analysis while maintaining 95%+ CSAT. Maintained zero critical compliance violations across audit cycles and sustained high-quality performance.

Operations Executive — Process & Documentation at SAR Group

January 1, 2019 - December 31, 2021

Managed end-to-end documentation workflows including MIS reports and structured operational feedback. Redesigned client onboarding checklists and SOPs to reduce onboarding time by 40%. Reduced manual data entry errors by 30% through systematic checklist implementation.

Customer Service Officer — Quality & Compliance at Home Credit India Pvt. Ltd.

January 1, 2018 - December 31, 2019

Delivered 98%+ call quality scores across monthly audits by adhering to strict compliance standards. Managed high-volume, quality-sensitive customer interactions with empathy and accuracy. Ranked in the top 5% of customer service officers regionally for two consecutive quarters.