Make yours proper
Bad-data days, hand-deployed pipelines, numbers nobody trusts. I re-architect what's there — tests and audits first — until the figure is quoted without a second look.
Data Platform Engineer · Analytics infrastructure
Platforms the business bets on.
I own analytics platforms end to end — ingestion, modeling, data quality gates, and the catalog that makes it all discoverable. Whether that means rebuilding a stack nobody trusts, or standing one up from empty object storage. ClickHouse, Airflow, dbt / SQLMesh, AWS.
Digital nomadGeorgia · UTC+4
NowData Platform Engineer, TKB Investment Partners
AvailableFull-time or contract, remote
Core stackClickHouse · dbt / SQLMesh · Airflow · AWS · Iceberg
Also buildingAn AI harness over the company's systems — one chat, everything reachable
Elsewheregithub.com/meteoFurletov
How I workGive me the problem and the keys. I'll come back with a platform.
Build it from scratch
Empty object storage to a working medallion lakehouse: ingestion, modeling, orchestration, catalog, CI/CD — infrastructure as code from the first commit.
Decide, then own it
I write the requirements as tests before anything else. The architecture usually becomes obvious after that.
Built for agents
A platform an agent can operate needs clean contracts, real lineage, and a catalog it can read. I build all three. I'm integrating Hermes, an open-source agent, as the agent managers work with: self-serve queries and Excel automation with no developer in the loop. Most of the code is written by Claude Code and Codex now. My job is deciding what's right and verifying it, which only works when the process is explicit, so I packaged mine as a Claude Code plugin: six stages, and a spec stage that is BDD, where every behavior becomes a Gherkin scenario the build is written against.
~5×Lower TCOServerless, scale-to-zero lakehouse, measured against the VM it replaced.
500Rows / secProduction ClickHouse market-analytics stack, stable throughput.
~0Data incidentsDown from one to two bad-data days a month, via SQLMesh audits. Held for a year.
minTo onboard a sourceStandardized pipeline deployment; it used to take hours.
Selected work
All work →Platforms, end to end
AWS lakehouse twin
The closing iteration of my serverless lakehouse: an AWS twin of the same platform, built from the same data and the same models. The Yandex Cloud original keeps running beside it. S3, Spark on AWS Glue, Iceberg tables, the Glue Data Catalog, Athena, Secrets Manager, Budgets and Cost Anomaly Detection, all in Terraform. I learned AWS by building this. The story is what changed moving from DuckDB and pointer files to Glue and Iceberg.
Read more →Serverless data lakehouse, four iterations
One analytics platform for a bookstore chain, rebuilt four times as the workload taught me what it needed. First PostgreSQL with Airflow in Docker on Yandex Cloud and DataLens dashboards. Then Airflow with SQLMesh models, four Postgres instances and Superset: a proper transformation layer, on a large always-on VM for a small workload. Then a serverless, scale-to-zero lakehouse: Parquet on object storage, DuckDB for every transformation, YAML data contracts that validate each dataset and generate the catalog, atomic publish, all Terraform, at about 5× lower TCO than the VM it replaced. The fourth iteration is the AWS twin. Finished in 2026; it runs unattended.
Read more →Company data platform, hybrid S3 + ClickHouse
Re-architected the company data platform on a hybrid S3 + ClickHouse medallion design, with analytical queries served directly on low-cost object storage. Turned a platform with one to two bad-data days a month into the trusted source of truth: SQLMesh audits and tests drove data-quality incidents to near zero, where they have held for a year. Standardized pipeline deployment, so onboarding a new source went from hours to minutes. Deployed OpenMetadata as the company-wide catalog and wrote the catalog standards that other teams now follow.
Read more →Company AI infrastructure
Open WebUI as the shared interface for open-source LLMs, vLLM for inference with observability, and a RAG system built for legal-document search. Internal services are getting MCP interfaces, with Hermes, an open-source agent, as the main agent analysts work with: one chat that reaches the whole stack.
sdlc-loop
The six-stage loop I work in, packaged as a Claude Code plugin so it is fixed in one place instead of re-derived in every repo: intent, spec, plan, build, review, watch. The spec stage is BDD. Every behavior becomes a Gherkin scenario in a .feature file, and that file is the contract the build is written against and reviewed against. A hook blocks edits to an existing scenario, so the target cannot move quietly to meet the code. Project-agnostic, with no language or stack assumption in it.
Hiring
I'm looking for a Data Platform Engineer role: remote, full-time or contract.
I build analytics platforms as systems: storage and engine, transformation and quality gates, catalog, infrastructure as code, and an AI layer on top. Digital nomad, based in Georgia, on UTC+4.
Start a conversationmeteofurletov@gmail.comlinkedin.com/in/meteofurletovgithub.com/meteoFurletov
BasedGeorgia · UTC+4
AvailabilityFull-time or contract