Dzmitry Dankov
I’m a Python-focused Data Engineer with about 6 years of experience designing and optimizing data pipelines for high-volume systems, including AI-enabled reporting.
About
I’m a Python-focused Data Engineer with about 6 years of experience designing and optimizing data pipelines for high-volume systems, including AI-enabled reporting. I’ve built real-time and batch ETL/ELT for IoT telemetry coming from 20+ vendors—handling thousands of sensors and sustaining hundreds of QPS—while keeping the end-to-end workflow reliable from ingestion to analytics and dashboards.
I also enjoy bridging engineering and product impact: I’ve migrated batch-only architectures toward lambda-style designs with event streaming, redesigned time-series data models to dramatically reduce storage (from 1.5 TB to 100 GB) without user-facing changes, and delivered reporting data marts in Snowflake. Alongside hands-on pipeline work, I’ve led backend/data engineering efforts, mentored teammates, and temporarily covered DevOps responsibilities when needed to keep deployments and operations smooth.
- 6 years' experience
- Joined October 2026
Skills· 20
Languages
- EnglishIntermediate
- RussianFluent
Industry experience
- Software & Internet
- Real Estate & Construction
- Telecommunications
- Gaming
- Professional Services
- Financial Services
- Travel & Hospitality
- Other
- Education
Work experience· 7 roles
Owned end-to-end ingestion, storage, and reporting for an IoT Smart Building analytics platform ingesting real-time telemetry from 20+ vendors (thousands of sensors, 300+ QPS sustained). Re-architected a batch-only ETL into a lambda architecture using an Azure Event Hubs speed layer and batch processing on SingleStore/PostgreSQL. Reduced production storage from 1.5 TB to 100 GB via time-series model redesign, retention policies, and aggregation strategy without user-facing impact. Designed time-series data models and rewrote analytical SQL to reduce dashboard/query latency. Built a Snowflake reporting data mart sourcing aggregates from operational stores. Implemented a historical backfill workflow with Apache Airflow + PySpark, including orchestrated retries/alerting and multi-month reprocessing. Added vendor-format validation with Pydantic and pipeline observability via Prometheus/DDTrace. Built async ingestion APIs and vendor integrations on FastAPI, and led a LangGraph multi-agent P
Developed backend and data layer for an HR/recruitment SaaS handling sensitive candidate data. Implemented a CDC pipeline (PostgreSQL → PGSync → OpenSearch) powering semantic search and deduplication over candidate profiles. Designed database architecture, ORM optimization, and Alembic migrations. Integrated third-party APIs and built real-time user activity synchronization using Centrifugo. Architected IAM infrastructure using Ory (Kratos, Oathkeeper, Keto) in Kubernetes and supported microservice integrations with deployments via Argo CD. Delivered features end-to-end in coordination with PM/BA/Frontend and DevOps.
Built the backend of a hotel WiFi billing and session-accounting system on top of a FreeRADIUS cluster. Implemented FreeRADIUS setup automation with Ansible, developed a custom RADIUS API, and built session accounting and billing data flows. Delivered customer authorization and admin tooling for WiFi session management, combining REST APIs and JWT-based interfaces, and supported operational reliability of the RADIUS-based stack.
Hire Dzmitry Dankov today
Post a job and invite Dzmitry, or contact them directly. No hidden commission.