Masum M.

Key Skills

Software

Google Cloud Vertex AI

Telus

Other

Internal/Proprietary Tooling

Top Subject Matter

HRL (Human-in-the-Loop) AI evaluation

Responsible AI

multilingual Indian language model fine-tuning.

Top Data Types

Computer Code Programming

Document

Text

Top Task Types

Computer Programming/Coding

Evaluation/Rating

Prompt + Response Writing (SFT)

Red Teaming

Fine-tuning

Freelancer Overview

AI Training & Systems Specialist (AI training, evaluation, and red teaming) at Uber Technologies Inc. (EXT).. Brings 8+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include Google Cloud Vertex AI and Other. Education includes Master of Science, Visva Bharati University (2019) and Bachelor of Science, Visva Bharati University (2017). AI-training focus includes data types such as Document and Text and labeling workflows including Red Teaming and Fine-tuning.

Labeling Experience

AI Training & Systems Specialist (AI training, evaluation, and red teaming) at Uber Technologies Inc. (EXT).

Google Cloud Vertex AIDocumentRed Teaming

Led red teaming and adversarial testing to identify LLM vulnerabilities and reduce edge-case failure rates for HRL-based AI products. Translated complex model and evaluation guidance into repeatable workflows and SOPs used for assessing instruction-following quality. Coordinated responsible AI prompt engineering for culture-sensitive use cases to improve alignment and reliability. • Conducted adversarial testing and purple teaming for model vulnerability discovery. • Developed SOPs for data evaluation workflows across diverse demographic and cultural contexts. • Evaluated fine-tuning outcomes for Indian languages (Bengali, Hindi, Indian English) using HRL-style assessment. • Managed cultural-based prompt engineering for the Nano Banana project and tracked alignment improvements.

2025 - Present

Data Annotator

TelusImageEvaluation/Rating

Performed image, video, and audio annotation according to client-specific guidelines, ensuring high-quality labeled data for model training. Contributed to training and refinement of AI models, including those involving sensitive and complex topics.

2024

Mathematics AI Model Trainer & SME at Chegg India Pvt. Ltd.

OtherTextFine-tuning

Trained and calibrated AI mathematical reasoning behavior using structured logic frameworks and guideline updates to support multi-path, non-binary judgment tasks. Developed step-by-step reasoning SOPs to help train AI models for complex logic deduction and higher accuracy. Maintained consistency across thousands of AI-generated educational responses by managing instructional policies and QA alignment. • Built instructional design and reasoning frameworks for advanced mathematics problem-solving. • Trained teams for nuanced, judgment-based tasks with multiple valid reasoning paths. • Updated instructional policies and performed QA alignment to ensure global consistency. • Supported model training using SOPs and guideline management for educational outputs.

2019 - 2025

Education

Work History