Md Osama Bin Mazhar

Data Scientist

India $30/hr Remote Full-time 9+ yrs exp 100 · Outstanding

About

Data Scientist with 5+ years of experience in machine learning, statistical analysis, data analysis, and AI research operations. Skilled in Exploratory Data Analysis (EDA), statistical analysis, feature engineering, model development, model validation, and production-oriented ML solutions. Experienced in supervised and unsupervised learning, predictive modeling, classification, segmentation, forecasting, A/B testing, and evaluation metric design. Strong command of Python, SQL, PyTorch, TensorFlow, Scikit-Learn, Pandas, NumPy, Docker, REST APIs, and cloud-based data science workflows. Proven ability to translate complex business problems into practical, data-driven solutions and deliver high-quality ML systems in remote research environments.

Skills & Expertise (69)

Python Advanced

9.0/10

Years Exp

SQL Advanced

8.5/10

Years Exp

Docker Advanced

8.5/10

Years Exp

PyTorch Advanced

8.5/10

Years Exp

TensorFlow Advanced

8.5/10

Years Exp

Data Cleaning Advanced

8.0/10

Years Exp

Exploratory data analysis Advanced

8.0/10

Years Exp

Statistical Analysis Advanced

8.0/10

Years Exp

Feature Engineering Advanced

8.0/10

Years Exp

Model Evaluation Advanced

8.0/10

Years Exp

VS Code WSL2 GitHub Actions Containerization Docker Compose Data Pipelines Prompt Engineering LLM Evaluation Data Preprocessing Data Extraction SQL Query Optimization Data Transformation Artifact Registry Debugging Code Review Performance Optimization FastAPI Flask API Development Matplotlib Seaborn Jupyter Notebook Report Generation stakeholder communication Git GitHub Docker Desktop LINUX Recommendation Systems JavaScript Bash PowerShell REST APIs YAML JSON Feature Selection A/B testing Experiment Design Predictive Modeling Classification Regression Clustering Segmentation Forecasting OpenpyXL Supervised Learning Unsupervised Learning Model Training Model validation Model Optimization Hyperparameter Tuning Cross-Validation Evaluation Metrics Anomaly Detection JAX scikit-learn Pandas NumPy SciPy

Work Experience

Software Engineer (Python & Docker)

Turing

Aug 2026 - Present

Designed, validated, and hardened data-driven evaluation tasks supporting LLM evaluation and AI research operations across the full task lifecycle. Built and debugged Docker-based evaluation environments with multi-service orchestration (Slack, Jira, Confluence, Zeta3 SQL, Google Drive, Freshdesk, Email). Performed exploratory data analysis (EDA) and statistical analysis on evaluation datasets to identify trends, patterns, and failure modes. Ran and analyzed 5x GLM-5.2 evaluation batteries, measuring pass rates against difficulty bands (0/5 to 2/5) and performing trajectory failure attribution. Defined evaluation metrics and assessed model performance across correctness, robustness, and policy compliance. Executed three-tier QC gates: Unified QC, Final QC, and Ship Delivery Gate, achieving 100/100 Ship QC on validated tasks. Produced structured evaluation artifacts (evaluations/ folder with Oracle runs, GLM accuracy runs, stability repeats) and wrote JSON verifier provenance. Debugged Unicode/encoding issues (cp1252 vs UTF-8) in Windows Python tooling and implemented PYTHONUTF8 workarounds. Used Google Cloud Shell, gcloud CLI, and Artifact Registry to pull private Docker images across environments.

Data Scientist / Machine Learning Engineer

Freelance

Jan 2020 - Present

Developed machine learning training, evaluation, and inference pipelines; automated data preprocessing, feature engineering, and dataset validation. Performed Exploratory Data Analysis (EDA) and statistical analysis to understand business and operational problems. Applied supervised and unsupervised learning techniques based on business requirements. Built predictive models for forecasting, classification, and segmentation use cases. Designed reusable benchmarking and evaluation frameworks for model performance and inference quality. Defined evaluation metrics and assessed model performance across multiple tasks. Worked extensively with Python, SQL, Pandas, NumPy, Scikit-Learn, TensorFlow, and PyTorch. Debugged and optimized production-style ML code for maintainability and performance.

LLM S2 Annotator (CUA Trajectory Specialist)

Turing

Mar 2026 - Jun 2026

Evaluated AI-generated trajectories involving multi-step reasoning, workflow execution, and tool usage for LLM research and benchmarking. Performed benchmark-driven evaluation, failure-mode analysis, and reasoning checks for correctness and policy compliance. Contributed to large-scale AI benchmarking and model evaluation projects, supporting research operations and quality assurance. Communicated analytical findings to technical and non-technical stakeholders through structured reports and recommendations.

AI Data Annotation Specialist

OpenTrain AI

Jan 2022 - Jan 2024

AI training data annotation, dataset validation, and QA; supported benchmarking through structured evaluation and feedback generation. Maintained annotation quality standards and contributed to research data preparation for AI/ML evaluation workflows. Produced preference data and evaluation signals used to improve model performance.

Education

Bachelor of Technology (B.Tech) in Computer Science and Engineering

2017 - 2021 · India

Certifications

Oracle Cloud Infrastructure 2025 Certified Generative AI Professional

Oracle University · 2025

Oracle Certified Professional certification covering Generative AI fundamentals, LLM architectures, prompt engineering, fine-tuning, RAG pipelines, and deploying Generative AI solutions on Oracle Cloud Infrastructure (OCI).

Oracle Cloud Infrastructure 2025 Certified Architect Associate

Oracle University · 2025

Oracle Certified Associate certification covering cloud architecture fundamentals on Oracle Cloud Infrastructure (OCI) — compute, storage, networking, identity, and security.