Founder and builder of MLdeck, focused on browser-local machine learning, privacy-aware AutoML, ONNX model workflows, WebAssembly, data analysis, and practical open-source engineering.

I bring more than 20 years of professional experience across IT, business intelligence, software development, data analysis, and systems architecture. My current work connects that experience with usable machine-learning products and open-source infrastructure.

• Founder and builder of MLdeck — browser-local, privacy-first AutoML for CSV-based machine-learning workflows.
• ONNX Member Company — MLdeck is listed on the official ONNX website as a Member Company. See the accepted member-company contribution.
• ONNX upstream contributor — authored the accepted Python 3.14 Pyodide wheel contribution for the official ONNX build and release workflow.
• Browser-local ONNX engineering — published a reproducible ONNX-for-Pyodide project with an end-to-end browser proof.

MLdeck helps users inspect CSV datasets, select features and targets, train and compare models, review baseline performance, identify data-quality and leakage risks, generate reports, and prepare validation-oriented export artifacts.

During normal browser training flows, raw CSV data is not uploaded to MLdeck servers. Exported artifacts should still be independently validated before use outside MLdeck.

• Website: https://mldeck.com/
• Public documentation: https://github.com/metalmancode/mldeck
• Browser-based AutoML: https://mldeck.com/browser-based-automl
• Privacy-first AutoML: https://mldeck.com/privacy-first-automl
• Local AutoML for CSV: https://mldeck.com/local-automl-csv
• Train machine-learning models in the browser: https://mldeck.com/train-ml-model-in-browser
• AutoML without uploading raw CSV data: https://mldeck.com/automl-without-uploading-data
• Export ONNX from the browser: https://mldeck.com/export-onnx-browser
• ONNX export workflow: https://mldeck.com/examples/onnx-export-workflow
• Local AutoML vs Cloud AutoML: https://mldeck.com/compare/local-automl-vs-cloud-automl

A reproducible WebAssembly build of the ONNX Python package for Pyodide, plus an end-to-end browser proof covering local model training, conversion with skl2onnx, ONNX graph validation and serialization, ONNX Runtime Web inference, and model download.

• Repository: https://github.com/metalmancode/onnx-pyodide
• Live browser proof: https://metalmancode.github.io/onnx-pyodide/demo/
• Official ONNX contribution: onnx/onnx#8192
• MLdeck member-company contribution: onnx/onnx.github.io#241

Public-safe product documentation, examples, architectural notes, validation limits, and browser-local AutoML workflows.

• Repository: https://github.com/metalmancode/mldeck
• Product: https://mldeck.com/

• Browser-local machine learning and privacy-aware AutoML
• ONNX model construction, export, validation, and runtime workflows
• Pyodide, WebAssembly, Python packaging, and cross-compilation
• Applied machine learning for tabular and CSV data
• Data preprocessing, feature engineering, and leakage-risk review
• Model evaluation, baseline comparison, explainability, and reporting
• Data quality, validation evidence, and portable model artifacts
• Business intelligence, KPI reporting, and data-product architecture
• AI-assisted software engineering and technical documentation

Machine Learning and Data: Python, Pandas, NumPy, Scikit-learn, Jupyter, classification, regression, exploratory data analysis, preprocessing, feature engineering, model evaluation, AutoML, and validation workflows.

Browser ML and Interoperability: ONNX, ONNX Runtime Web, Pyodide, WebAssembly, skl2onnx, browser-local execution, and portable model artifacts.

Programming and Web: TypeScript, JavaScript, SQL, HTML/CSS, software architecture, APIs, testing, and technical documentation.

BI and Data Platforms: Power BI, Looker Studio, BigQuery, Excel, Google Sheets, dashboards, ETL/ELT concepts, KPI analysis, and business data modeling.

Cloud and IT Systems: Microsoft Azure, Azure Data Fundamentals, Power Platform, Office 365, SharePoint, IT architecture, infrastructure administration, and process optimization.

• SpaceX Falcon 9 Landing Prediction — classification-based IBM Data Science capstone project.
• Australia Rainfall Prediction — Scikit-learn pipelines, model comparison, and hyperparameter search.
• Mushroom Classifier — safety-focused classification with asymmetric error considerations.
• Bee-Haven Azure Lakehouse — Azure data lakehouse architecture for apiculture analytics.
• GoExplore Product Analysis — BigQuery, Looker Studio, and data-quality analysis.
• RAG Project — retrieval-augmented generation experiment.

• IBM Data Science Professional Certificate
• Machine Learning with Python — IBM / Coursera
• Data Analysis with Python — IBM / Coursera
• Data Visualization with Python — IBM / Coursera
• Generative AI and Prompt Engineering — IBM / Coursera
• Microsoft Certified: Azure Data Fundamentals
• PCEP — Certified Entry-Level Python Programmer
• Google IT Support / System Administration / IT Security
• Cisco Cybersecurity Essentials
• German B2 — telc
• English C1 — BA in English

• LinkedIn: https://www.linkedin.com/in/reza-r-759803171/
• GitHub: https://github.com/metalmancode
• MLdeck: https://mldeck.com/
• MLdeck public documentation: https://github.com/metalmancode/mldeck
• Collaboration: contact@mldeck.com

Building practical, privacy-aware machine-learning products and contributing the underlying engineering back to the open-source community.