Aleksandr Andreev

Lead Data Engineer building dependable platforms for data-intensive teams.

I design streaming and lakehouse systems, ship developer tooling with LLMs, and write about the trade-offs that matter once systems reach real scale.

What I usually bring

Production data platforms with explicit ownership and operability.

Streaming and lakehouse decisions grounded in cost, latency and team constraints.

Internal AI tooling that helps engineers move faster without hiding the system.

Featured writing

Apr 30, 2026

First post on the new site — stack overview and what's coming.

Active

LLM-powered code review bot using FastAPI, Qdrant, Claude API, and RAG over Confluence. Reduced review turnaround by 40%.

Apr 29, 2026

DuckDB pushes column projection and filter predicates down to S3 Parquet reads — but only if your file stats are valid.

Get new essays by email

Low-volume updates about data engineering, architecture and useful experiments.