Shamji Shekhava

Detail-oriented Transcriptionist, Data Annotator, and AI Content Evaluator with proven expertise in Hindi, Gujarati, and English language projects. Experienced in delivering high-accuracy transcriptions, audio evaluations, image annotations, and content safety assessments for AI training. Skilled in applying complex guidelines, maintaining quality standards, and working across multiple platforms including Lighthouse, OneForma, Outlier, Alignerr, and Sigma AI.

Seeking to leverage multilingual proficiency and technical skills in AI data processing and content evaluation roles.

Transcription & Recording Specialist at Joshtalk

Transcription

• Transcribed 500+ hours of Hindi/Gujarati audio (interviews, narratives, meetings) with 99% accuracy.
• Performed speaker identification, timecoding, and context-specific formatting.
• Produced studio-quality voice recordings; optimized audio clarity through noise reduction and vocal modulation.
• Proofread and edited transcripts to ensure linguistic precision and adherence to client style guidelines.

Transcription at Sigma AI (Spain, Madrid)

• Processed and transcribed 30+ hours of multilingual audio in Hindi with Maintaining accuracy.
• Managed end-to-end transcription workflow including speaker diarization, labeling, attribution, and precise timecoding.
• Applied context-aware formatting tailored to content type (interviews, narratives, business meetings).
• Conducted comprehensive proofreading and linguistic editing to ensure textual precision, cultural nuance, and readability.
• Gained hands-on experience with secure data compute platforms for large-scale audio processing, file transfer, and workflow management.
• Maintained data integrity throughout the project lifecycle, ensuring high-quality deliverables for AI training datasets.

Content Annotator & Evaluator at Lighthouse / Cherry Opal (Centific)

• Evaluated AI-generated content across multiple workflows:
• Preference Ranking (PR): Assessed and ranked AI responses based on instruction following, truthfulness, localization, and concision.
• AFM Safety Grading: Evaluated content for safety, harmfulness, and appropriateness.
• TA / WT Production: Graded AI-generated text responses for Personalized Smart Replies, Mail Smart Replies, and Intelligent Polls.
• CYU Summarization: Evaluated topic summarization and action item extraction.

12th Pass – GSHEB Gandhinagar (2017-03)

BA in Human Psychology (2024-07)