Home Projects Portfolio Dashboard Export PDF Log in

Languages

English
110

Latest Updates

Documenting code, one commit at a time.

Python 4 posts
×
0 Jupyter Python

Refining Data Workflows with Jupyter Notebooks

Introduction

The Helper-IA project focuses on building intelligent tools that simplify complex data processing tasks. Recently, I have been working on refining the internal workflows within our Jupyter notebooks to ensure that data transformations are modular, reproducible, and easier to maintain.

The Problem

As our data analysis notebooks grew in complexity, we encountered several

0 Jupyter Python

Streamlining Data Analysis with Helper-IA and Google Colab

The Challenge

When starting a new data-driven project, the initial hurdle is often setting up an environment that is both collaborative and computationally efficient. In the Helper-IA project, the goal was to standardize the Exploratory Data Analysis (EDA) process, ensuring that insights could be generated rapidly without the overhead of local configuration.

The Approach

Streamlining NL2SQL Pipelines with PostgreSQL and Supabase

When working on VeriQuery, our team aimed to enhance the natural language to SQL (NL2SQL) pipeline. The original pipeline encountered compatibility issues when used with PostgreSQL and Supabase. This post outlines the steps taken to rectify these issues and create a more robust and versatile system.

The Challenge

The initial NL2SQL pipeline was not fully compatible with PostgreSQL and

Enhancing Data Processing Pipelines with Robust Error Handling in VeriQuery

Introduction

Data processing pipelines are susceptible to errors, especially when dealing with external or dynamically changing data sources. In VeriQuery, ensuring the pipeline's resilience to unexpected data, like empty schemas, is crucial for maintaining reliable query generation. This post delves into how to enhance pipeline robustness by implementing proper error handling and logging.