Data Engineering

Oh That Data Girl

Sign in Subscribe

Latest — Dec 1, 2025

Build in Public: Week 4. The Messy Middle Of Building An AI Agent

This was supposed to be the week of a polished demo video and a clean “here’s how you use our API” walkthrough. Instead, it turned into the week of staring at half-finished pieces, poking at logs and wondering why we voluntarily chose Instagram as our first supported social network.

More issues

Build in Public: Week 3. First Survive Discovery, Then Enjoy Analysis

Last week I noticed something annoying: the engagement on my Week 1 and Week 2 posts dropped, even though the content was objectively good. So I asked Perplexity when developers actually read dev.to and the answer was basically: please stop posting on Saturdays. No one is there. From there,

Build in Public: Week 2. How Do People Even Find Influencers?

If you remember from the last update, we had our first real conflict: Node.js vs Python. Well, democracy has spoken. My "many" LinkedIn followers voted and the winner is Node.js. So this week we’re continuing with one backend, one direction and slightly fewer arguments. But

Build in Public: Week 1

The first week wasn’t particularly shiny. We mostly tried to move from our old n8n setup to a normal backend and figure out what “normal” even means for us. Somewhere along the way we also had our first team conflict and our first real issue with Bright Data, so

Build in Public: Day Zero

A few weeks ago we built an agent called WYKRA for the Real-Time AI Agents challenge powered by Bright Data and n8n. It was a small experiment that could accept a natural-language query like “I run a small bakery in Barcelona, find influencers who talk about bread here” and return

Three Formats Walk into a Lakehouse: Iceberg, Delta and Hudi in a Local Setup You Can Run on Your Laptop

I still remember building an entire bookkeeping system in FoxPro for my university coursework back in 2007. Planning every table structure upfront, carefully designing indexes because adding them later meant locking everything up for a reindex, managing backups manually, worrying about file corruption. Every schema change was a small crisis.

When Small Parquet Files Become a Big Problem (and How I Ended Up Writing a Compactor in PyArrow)

It all began with a fairly normal data pipeline. Events were coming in through Kafka, landing in AWS S3 as Parquet files after going through some lightweight microbatch processing. It looked clean at first glance. Efficient. Predictable. But one day I opened one of the hourly folders and saw the

You’re Not the First. You Won’t Be the Last. But We Survive

A WeCoded 2025 Story. I didn’t grow up dreaming of being in tech - it just happened, one decision at a time. I was born in a provincial city in Belarus in a family of an engineer and an economist. We lived in a marginal neighborhood - not in

Who Wins Dev.to Challenges? (And The Hidden Benefits No One Talks About)

At the beginning of this year I made a decision - I was going to participate in every Dev.to challenge. Well, almost every challenge. Let’s be honest, frontend isn’t my thing, so I happily skipped those. But everything else? Fair game. Why? Because after writing on Dev.

Scaling Data Analytics: Building a Starter Kit with Neon, Airflow, and Streamlit

Setting up a data analytics project from scratch can be a headache. You need a database, a way to pull in data, and a dashboard to make sense of it all. I wanted to make that process easier—something lightweight, flexible, and beginner-friendly. That’s how I ended up building

About

Oh That Data Girl

Anti-Bullshit Enthusiast. Data engineer with a systems mindset, mildly owned by cats. Writing what I wish someone had written earlier.

Topics

blogging

dataengineering

Golang

ml

python

tutorial