I'm a Python Data Workflow Developer building small, reproducible projects for API data extraction, CSV/Excel data cleaning, JSON transformation, ETL, data validation, and reporting automation.

My current focus is helping small teams turn messy API, CSV, Excel, and JSON data into clean, validated, reporting-ready workflows.

• API / CSV / Excel / JSON data input
• data cleaning, validation, and transformation
• JSON flattening for reporting-ready tables
• data quality checks for handoff and review
• CSV, SQLite, PostgreSQL, and Parquet output
• PostgreSQL reporting tables and lightweight SQL views
• BI-ready handoff notes for local reporting workflows
• AI-ready data handoff artifacts without calling LLM APIs
• lightweight FastAPI data validation services
• pytest-based project checks
• Docker-based local development
• clear README files, screenshots, limitations, and usage notes

A small Python data quality ETL starter for cleaning, validating, exporting, and reporting messy CSV, Excel, JSON, and mock API-style data.

This is my main portfolio project for Python data workflow development.

It demonstrates a repeatable data workflow:

messy CSV / Excel / JSON / mock API-style data
        ↓
read and flatten
        ↓
normalize columns
        ↓
validate expected schema rules
        ↓
clean duplicate rows and text values
        ↓
export cleaned CSV + SQLite / optional PostgreSQL
        ↓
generate data quality reports

It also includes an optional analytics-ready path:

generated messy order data
        ↓
existing validation and cleaning logic
        ↓
cleaned CSV
        ↓
cleaned Parquet
        ↓
DuckDB query demo
        ↓
summary CSV tables + benchmark report

It also includes an optional BI-ready path:

generated messy order data
        ↓
validation and cleaning workflow
        ↓
analytics-ready order rows
        ↓
PostgreSQL reporting tables
        ↓
lightweight SQL views
        ↓
optional Metabase local dashboard demo

It also includes an optional AI-ready data preparation path:

generated messy order data
        ↓
validation and cleaning workflow
        ↓
cleaned orders dataset
        ↓
schema profile JSON
        ↓
data dictionary JSON
        ↓
validation summary JSON
        ↓
feature-ready CSV
        ↓
embedding-ready text field extract
        ↓
AI-ready manifest + Markdown summary report

What it shows:

• CSV, Excel, JSON, and mock API-style input
• nested JSON flattening
• column normalization
• Pydantic-based workflow and schema models
• missing value, duplicate, email, date, and number validation
• cleaned CSV output
• SQLite output by default
• optional PostgreSQL export
• optional FastAPI validation service
• optional generated order data for larger workflow demos
• optional analytics-ready CSV and Parquet export
• optional DuckDB local analytics query demo
• optional PostgreSQL reporting tables and SQL views
• optional Metabase local dashboard demo
• optional AI-ready data preparation artifacts
• schema profile JSON
• data dictionary JSON
• validation summary JSON
• feature-ready CSV output
• embedding-ready text field extract
• AI-ready manifest and Markdown summary report
• Markdown and JSON data quality reports
• lightweight summary tables
• benchmark report and BI summary report
• CLI execution
• pytest tests
• Docker-based execution
• practical documentation and screenshots

Repo: https://github.com/OnerGit/data-quality-etl-starter

Related writing:

• Preparing AI-Ready Data Without Calling an LLM API https://dev.to/bob_oner/preparing-ai-ready-data-without-calling-an-llm-api-5daf

• From Clean Data to BI-Ready Reporting Tables with Python, PostgreSQL, and Metabase https://dev.to/bob_oner/from-clean-data-to-bi-ready-reporting-tables-with-python-postgresql-and-metabase-348p

• From Data Quality Checks to Analytics-Ready Parquet with Python https://dev.to/bob_oner/from-data-quality-checks-to-analytics-ready-parquet-with-python-39bd

• Build a Python Data Quality ETL Starter for Messy CSV, Excel, JSON, and API-Style Data https://dev.to/bob_oner/build-a-python-data-quality-etl-starter-for-messy-csv-excel-json-and-api-style-data-46b

A lightweight FastAPI service that accepts CSV uploads and returns structured JSON data quality reports.

This project demonstrates how a data validation workflow can be packaged as a small API service.

What it shows:

• CSV upload handling
• pandas-based data quality checks
• row count and column count analysis
• missing value checks
• duplicate row detection
• empty column detection
• optional expected-column validation
• Pydantic response models
• structured error handling
• pytest coverage
• Docker packaging
• Swagger UI screenshots and documentation

Repo: https://github.com/OnerGit/fastapi-csv-quality-api

Related writing:

• Build a CSV Data Quality API with FastAPI, Pandas, Pytest, and Docker https://dev.to/bob_oner/build-a-csv-data-quality-api-with-fastapi-pandas-pytest-and-docker-28ld

A privacy-first Tampermonkey userscript for navigating long ChatGPT conversations locally in the browser.

This is not my main data workflow direction, but it demonstrates my engineering habits: small scope, local-first design, readable implementation, privacy boundaries, manual testing, documentation, screenshots, and clear limitations.

What it shows:

• browser-side UI enhancement
• collapse / expand controls
• MutationObserver support
• localStorage-based UI state
• no external API calls
• no telemetry
• no content upload
• practical troubleshooting notes

Repo: https://github.com/OnerGit/ChatGPT-Long-Conversation-Helper

Related writing:

• Build a Privacy-First Tampermonkey Script for Long ChatGPT Conversations https://dev.to/bob_oner/build-a-privacy-first-tampermonkey-script-for-long-chatgpt-conversations-2765

• Preparing AI-Ready Data Without Calling an LLM API https://dev.to/bob_oner/preparing-ai-ready-data-without-calling-an-llm-api-5daf

• From Clean Data to BI-Ready Reporting Tables with Python, PostgreSQL, and Metabase https://dev.to/bob_oner/from-clean-data-to-bi-ready-reporting-tables-with-python-postgresql-and-metabase-348p

• From Data Quality Checks to Analytics-Ready Parquet with Python https://dev.to/bob_oner/from-data-quality-checks-to-analytics-ready-parquet-with-python-39bd

• Build a Python Data Quality ETL Starter for Messy CSV, Excel, JSON, and API-Style Data https://dev.to/bob_oner/build-a-python-data-quality-etl-starter-for-messy-csv-excel-json-and-api-style-data-46b

• Build a CSV Data Quality API with FastAPI, Pandas, Pytest, and Docker https://dev.to/bob_oner/build-a-csv-data-quality-api-with-fastapi-pandas-pytest-and-docker-28ld

• Build a Privacy-First Tampermonkey Script for Long ChatGPT Conversations https://dev.to/bob_oner/build-a-privacy-first-tampermonkey-script-for-long-chatgpt-conversations-2765

• AI-Assisted Development Is Not Autopilot https://dev.to/bob_oner/ai-assisted-development-is-not-autopilot-15ie

I prefer small, practical engineering projects that are:

• runnable locally
• easy to test
• clearly documented
• honest about limitations
• based on realistic workflow problems
• structured for handoff and maintenance
• simple enough for a client or reviewer to inspect

I do not try to present every project as production infrastructure. I focus on clear, reviewable starter projects that can be adapted into client-specific workflows.

My current portfolio is most relevant to projects such as:

• cleaning messy CSV or Excel files
• validating required columns before reporting
• converting nested JSON into tabular data
• turning API-style data into CSV / Excel-ready outputs
• building repeatable Python ETL scripts
• generating data quality reports
• exporting cleaned data to SQLite or PostgreSQL
• preparing cleaned CSV / Parquet files for local analytics
• creating lightweight summary tables for reporting
• preparing PostgreSQL reporting tables and SQL views
• creating BI-ready handoff notes for local reporting workflows
• preparing schema profiles and data dictionaries
• creating validation summaries for downstream review
• preparing feature-ready CSV files
• extracting embedding-ready text fields without generating embeddings
• creating AI-ready data manifests without calling LLM APIs
• packaging a data validation workflow as a small FastAPI service
• adding tests, documentation, and handoff notes to existing scripts

My current portfolio is intentionally not focused on:

• enterprise data platforms
• production data warehouses
• Airflow / dbt / Spark pipelines
• Snowflake / Databricks projects
• full-stack SaaS applications
• production BI platform management
• Metabase production deployment
• LLM applications
• LLM-based data cleaning tools
• embedding generation pipelines
• vector database projects
• machine learning training pipelines
• prompt engineering frameworks
• RAG chatbots
• AI agents

The current goal is narrower: small, runnable, testable Python data workflows for cleaning, validation, export, reporting preparation, AI-ready data handoff, and documentation.

Python, pandas, FastAPI, Pydantic, pytest, Docker, API integration, CSV/Excel processing, JSON flattening, data validation, ETL, SQLite, PostgreSQL, Parquet, DuckDB, SQL views, reporting automation, schema profiling, data dictionaries, validation summaries, browser userscripts, and technical documentation.