Greetings to all curious minds!

I'm Akzhan — a Data & AI Team Lead based in Almaty, Kazakhstan, and this is Bad Dog Data.

I started this blog in 2023 when I was a Data Product Manager at a marketplace, mostly writing for myself and anyone curious enough to read about statistics, engineering, and the occasional rabbit hole. The posts were technical, a bit rough, and genuinely fun to write. That part hasn't changed.

Today, I lead a data team dedicated to cross-border e-commerce. My day-to-day is spent building the middleware and intelligence layers that connect massive international marketplaces to local platforms. We handle heavy market data ingestion, pull and normalize complex partner feeds, and build RAG (Retrieval-Augmented Generation) systems for dynamic resource localization. We also design and implement dynamic pricing pipelines that react in real-time to global market signals.

I also write open source. Right now, I'm implementing my library, lyubishchev, directly into our day-to-day production workflows. The R package is already out in the wild, and I'm actively preparing to propose the Python version for inclusion in SciPy and scikit-learn.

This blog still lives at the intersection of data, business, and the human experience. Some posts will be deep technical dives into vector search layers, dynamic pricing models, or normalization pipelines. Others will be about what actually happens when you try to build resilient data systems across borders with real constraints, messy APIs, and real politics.

If you're an engineer or analyst — welcome, there's plenty here for you. If you're a product owner or decision-maker trying to figure out how to leverage RAG, complex partner data, or automated pricing for your business — you might find this useful too.

Yours,
Bad Dog.