Deependra Singh

Key Skills

Software

Mercor

Top Subject Matter

Physics/Numerical Methods/Scientific Computing

AI Safety/Language Models

Speech Recognition/Language Data

Top Data Types

Text

Audio

Image

Top Task Types

RLHF

Audio Recording

Freelancer Overview

Physics Python Coding Expert (Independent Contractor). Brings 5+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include Mercor. Education includes Master of Science, NISER Bhubaneswar, Homi Bhabha National Institute (2025) and High School Diploma, St. John’s School Marhauli (2020). AI-training focus includes data types such as Computer Code, Programming, and Text and labeling workflows including Evaluation, Rating, and RLHF.

Labeling Experience

Hindi Voice Actor (Independent Contractor)

MercorAudioAudio Recording

Contributed high-quality Hindi voice recordings for AI training datasets, focusing on accurate pronunciation and natural delivery. Tasks required linguistic precision and meeting clarity standards for automated speech recognition models. The experience enhanced dataset diversity and model performance in multilingual contexts. • Recorded conversational and scripted phrases • Verified linguistic and acoustic accuracy • Ensured recordings met AI training requirements • Addressed language diversity for AI model robustness

Contributed high-quality Hindi voice recordings for AI training datasets, focusing on accurate pronunciation and natural delivery. Tasks required linguistic precision and meeting clarity standards for automated speech recognition models. The experience enhanced dataset diversity and model performance in multilingual contexts. • Recorded conversational and scripted phrases • Verified linguistic and acoustic accuracy • Ensured recordings met AI training requirements • Addressed language diversity for AI model robustness

2026 - 2026

RLHF Safety & Adversarial Generalist (Independent Contractor)

MercorTextRLHF

Evaluated and stress-tested large language model outputs for safety, robustness, and alignment. Responsibilities included adversarial prompt design, analyzing model responses for safety risks, and documenting edge cases for model improvement. The role required deep understanding of natural language and AI safety benchmarks. • Focused on adversarial testing for model vulnerabilities • Rated responses for appropriateness and safety • Provided detailed alignment feedback to developers • Enhanced model reliability through iterative RLHF tasks

Evaluated and stress-tested large language model outputs for safety, robustness, and alignment. Responsibilities included adversarial prompt design, analyzing model responses for safety risks, and documenting edge cases for model improvement. The role required deep understanding of natural language and AI safety benchmarks. • Focused on adversarial testing for model vulnerabilities • Rated responses for appropriateness and safety • Provided detailed alignment feedback to developers • Enhanced model reliability through iterative RLHF tasks

2026 - 2026

Physics Python Coding Expert (Independent Contractor)

Mercor

Reviewed and validated AI-generated solutions for advanced physics problems to ensure correctness and scientific validity. Designed novel, challenging problem statements to test and expose failure modes in state-of-the-art language models, focusing on physics and numerical methods. This involved detailed assessment, iterative refinement, and in-depth scientific understanding alongside AI data evaluation. • Evaluated Python code and scientific computing AI responses • Flagged errors and provided feedback for model improvement • Designed and analyzed advanced algorithmic prompts • Improved AI model robustness through targeted data labeling tasks

Reviewed and validated AI-generated solutions for advanced physics problems to ensure correctness and scientific validity. Designed novel, challenging problem statements to test and expose failure modes in state-of-the-art language models, focusing on physics and numerical methods. This involved detailed assessment, iterative refinement, and in-depth scientific understanding alongside AI data evaluation. • Evaluated Python code and scientific computing AI responses • Flagged errors and provided feedback for model improvement • Designed and analyzed advanced algorithmic prompts • Improved AI model robustness through targeted data labeling tasks

2026 - 2026

Education

Work History