Developed and implemented a modern data platform to get actionable business insights from a fictional pizza store's day-to-day sales data. The project leverages an open-source modern data stack, incorporating tools like Apache Airflow for orchestration, dbt Core for data transformation, BigQuery as the data warehouse, Python-based jobs for data processing, and Lightdash for analytics and visualization.
(click for more details...)
In this project, I have integrated a Large Language Model (LLM) with a custom-built application to enhance its functionality. The LLM is used to process and analyze user inputs, providing intelligent responses and insights. The integration allows the application to leverage the power of natural language processing, enabling it to understand and generate human-like text. Link to
Medium Article
for more details on the project.
MockDataGen is a Python-based utility (PyPi) designed to generate synthetic data for testing and development purposes. It allows users to create realistic datasets that mimic real-world data structures, enabling developers to test applications without relying on sensitive or proprietary information. The tool supports various data types and formats, making it versatile for different use cases.
(click for more details...)
End to End data pipeline extracting my Strava Acitivities. Showcasing data extraction from the Strava API into BigQuery cloud. Once ingested, BigQuery pipelines handle data cleanup and transformation. These cleaned and enriched datasets are then published to Looker Studio, where users can explore their activities through interactive BI Visuals. The result is a fully automated, end-to-end analytics pipeline—from raw API to polished BI.
(click for more details...)
A lightweight Python package (PyPi) that centralizes database connection management for data engineering workflows. It uses a YAML-based interface to handle credential storage, lets you add/update/test connections via CLI or code, and returns ready-to-use connectors for Oracle, MySQL, SQL Server, and PostgreSQL with a single import. Whether you're prototyping ETL scripts or deploying production pipelines, this tool removes boilerplate and gives you reliable DB access on demand.
(click for more details...)
Designing a robust data model is often a iterative process and is implemented in layers (Conceptual/Logical/Physical Data Model). The aim of this project is to cutdown on this effort and allow fast protyping, also allowing a Business user to create Data Models from scratch without understanding the knowhow of RDBMS systems (click for more details...)
This utility allows user to run easy to remember SQL code's Ex- heck Database Online or Describe table and takes care of running complex queries for choosen RDBMS in the background.. (click for more details...)