# Ivan Blinov

> Hands-on data and AI platform architect. I build data platforms where the numbers have to be right to the cent, and the AI-agent infrastructure that lets a small team run them safely. Tech lead for the billing data of millions of sellers at one of the largest marketplaces in Eurasia (100+ TB); co-founder and CTO of a health-tech startup in Brazil; earlier an engineering manager whose team grew from 6 to 20. I work with coding agents and verify everything they produce.

Data & AI Platform Architect, works remotely across the Americas (UTC-3 to UTC-6). Languages: Russian (Native), English (C2), French (C2), Spanish (C2), Portuguese (B2). Contact: contact@ivanblinov.com

Every page on this site is also Markdown at the same URL: curl gets it by default, other clients send `Accept: text/markdown` or append `index.html.md` to the URL. A read-only MCP server with the same content runs at https://ivanblinov.com/mcp (Streamable HTTP, protocol revision 2026-07-28, with `initialize` for 2025 clients).

## Profile
- [About](https://ivanblinov.com/about/index.html.md): background, how I work, languages, contact
- [CV](https://ivanblinov.com/cv/index.html.md): the full CV
- [resume.json](https://ivanblinov.com/resume.json): the profile in JSON Resume format

## Case studies
- [AI agents for a data team, without handing them the keys to production](https://ivanblinov.com/writing/ai-agents-for-a-data-team/index.html.md): What I built in 2026 for a 12-engineer billing data team, what the agentic code reviewer actually caught, and which guardrails mattered.
- [Idle CPUs, 60-second queries: a global lock in ClickHouse 23.8](https://ivanblinov.com/writing/clickhouse-global-context-lock/index.html.md): How a dashboard widget that needed two numbers took a ClickHouse cluster down for 48 minutes, and why the fix was a cheaper query rather than more hardware.
- [665 GiB of dictionaries: when ClickHouse dictionaries stop scaling](https://ivanblinov.com/writing/clickhouse-dictionaries-665-gib/index.html.md): Nine hashed dictionaries quietly grew to two thirds of a terabyte per node, and every reload tried to build a second copy. How we found that only five views used them, and replaced them with joins that read 17 granules out of 282,102.
- [Airflow tasks stuck in "queued" forever after a twelve-second database blip](https://ivanblinov.com/writing/airflow-tasks-stuck-in-queued/index.html.md): A short network flap between two datacenters left a task in "queued" for two and a half hours, and its DAG with it. The cause is a lost executor event: `SELECT … FOR UPDATE SKIP LOCKED` skips exactly the row it needs. Why restarting the scheduler is the wrong fix.
- [Expired certificates broke production twice, and nothing turned red](https://ivanblinov.com/writing/expired-certificates-twice/index.html.md): Once as a load-balancer black hole that swallowed 23 of 24 connections for fifteen hours, once as a silent 30-hour replica lag behind a database that kept answering. Both times the metrics existed and the alerts did not.

## Notes
- [LowCardinality in ClickHouse: I measured the rule of thumb, and it was wrong both ways](https://ivanblinov.com/writing/clickhouse-lowcardinality-measured/index.html.md): "LowCardinality helps between a thousand and a hundred thousand distinct values, and hurts otherwise." A code review quoted this rule, so I measured it on ClickHouse 23.8. LowCardinality wins at any small number of values, even two, and starts losing badly at about twenty thousand per part.
- [Partitioning by modulo in ClickHouse: the DDL I wrote that made queries 13.7 times slower](https://ivanblinov.com/writing/clickhouse-modulo-partitioning/index.html.md): `PARTITION BY supplier_id % 100` sounds like it spreads the load evenly. In practice it destroyed primary-key pruning: a lookup read 64% of the table instead of 3%. Here are the measurements, and why it happens.

## For agents
- [For agents](https://ivanblinov.com/agents/index.html.md): machine interfaces, the MCP server and etiquette

## Optional
- [llms-full.txt](https://ivanblinov.com/llms-full.txt): every page above in one file
- [Atom feed](https://ivanblinov.com/feed.xml): new case studies and notes