Sourav Pal

x

AI Language Specialist Tutor

xAI (Grok)

Dec 2025 - Current (6 months)

• Generated and annotated high-quality bilingual (Hindi & English) datasets to support training of advanced AI models, including systems aligned with Grok's conversational intelligence and personality. • Evaluated and reviewed large volumes of text and audio data for linguistic accuracy, clarity, cultural relevance, and contextual appropriateness. • Contributed to AI model improvement in social cognition, emotional intelligence (EQ), truthfulness, and conversational style dimensions. • Identified and categorized errors including grammar, syntax, semantic inconsistencies, transcription issues, and non-native phrasing patterns. • Created and refined multi-modal content (text, voice, and video) to enhance model training and real-world applicab

Linguistics Expert - Freelance AI Trainer

Innodata Inc.

Nov 2025 - Feb 2026 (4 months)

• Crafted clear, fact-checked, and structured responses to train AI language models, ensuring accuracy, coherence, and linguistic quality. • Designed and generated challenging prompts to improve AI reasoning, logic, and contextual understanding. • Developed and applied evaluation criteria and scoring rubrics to assess AI outputs for accuracy, relevance, and consistency. • Performed data annotation and content review to enhance NLP and machine learning model capabilities.

M

Domain Expert (SME) / LLM Trainer – Mathematics

Mindrift Al

Jun 2025 - Current (1 year)

• Evaluated and fine-tuned large language models (LLMs) to improve mathematical reasoning across Algebra, Calculus, and Probability domains. • Developed high-quality step-by-step solutions using Chain-of-Thought (CoT) methodology and complex problem sets to enhance AI accuracy, logic, and multi-step reasoning. • Performed quality assurance and validation of AI-generated mathematical content, ensuring correctness, consistency, and adherence to academic standards. • Created and structured LaTeX-formatted datasets for training and evaluating machine learning and NLP models. • Designed curriculum-aligned educational content and assessments supporting AI-driven learning tools.

C

AI Translation Rater

CrowdGen by Appen

Jan 2025 - Dec 2025 (1 year)

• Rated AI-generated translations for accuracy, fluency, and cultural appropriateness against reference examples. • Corrected incorrect or awkward translations, rewriting to ensure natural-sounding, contextually accurate output. • Assigned semantic similarity scores to evaluate alignment of machine translations with original intent. • Flagged harmful, offensive, or inappropriate content for content moderation and safety compliance.

Data Analyst - AI & Search Quality

Peroptyx

Apr 2024 - Current (2 years 2 months)

• Conducted comprehensive data validation and quality assurance for machine learning training datasets. • Evaluated and validated accuracy of ML data outputs contributing directly to AI model improvement cycles. • Performed detailed analysis of search results including news, images, and transit data for relevance verification. • Reviewed and authenticated mapping data for digital navigation applications ensuring accuracy and safety. • Applied advanced research methodologies to cross-check AI responses and verify factual accuracy.

Generative AI Trainer & Quality Lead

OneForma

Nov 2023 - Current (2 years 7 months)

• Delivered training on Generative AI, Large Language Models (LLMs), and prompt engineering, enabling contributors to produce high-quality AI training data. • Led quality assurance and evaluation of AI-generated outputs for accuracy, relevance, bias detection, and compliance with project guidelines. • Designed and maintained SOPs, training materials, and evaluation rubrics to standardize annotation workflows and improve model performance. • Labeled and validated PII data per project-specific requirements and data governance standards. • Mentored and onboarded contributors, providing actionable feedback to enhance productivity and task accuracy. • Collaborated with PMs, QA teams, and cross-functional stakeholders via Microsoft Teams to achie

Google Gemini LLM Data Rater

RWS Group

Aug 2023 - Current (2 years 10 months)

• Evaluated multi-modal content (text, images, audio, video) to ensure search relevance, accuracy, and user intent alignment for Google Gemini AI. • Conducted query intent analysis and flagged low-quality or misleading results to improve search quality. • Annotated and curated large-scale datasets supporting AI training, machine learning, and NLP development. • Performed model evaluation and quality checks ensuring AI outputs were accurate, unbiased, and safe.

T

Online Data Analyst - Search Quality

Telus International (TELUS AI)

Dec 2021 - Jan 2023 (1 year 2 months)

• Evaluated and rated search engine results for relevance, accuracy, and user intent per TELUS AI guidelines. • Analyzed queries to identify informational, navigational, and transactional user intent to improve SERP quality. • Conducted web research and data validation to ensure accuracy and credibility of web content and multimedia. • Contributed to AI training models by providing structured feedback and high-quality data annotations.

N

AI Transcription Quality Analyst (Hindi)

Netscribes Pvt. Ltd.

Feb 2021 - Jun 2021 (5 months)

• Reviewed and validated AI-generated Hindi audio transcriptions for accuracy and completeness. • Compared audio recordings with transcripts to identify and correct errors in grammar, spelling, and context. • Interpreted various Hindi accents, dialects, and speech patterns to ensure transcription consistency. • Provided structured feedback to improve AI transcription performance and overall quality metrics.

About

Detail

Timeline

Résumé