Anton Vlasenko
Data engineering lead Systems architect
I build data platforms and ML pipelines, then run them in production.
- 10 years
- 50+ projects
- 50 TB a month, largest pipeline
- 4 industries: logistics, banking, healthcare, automotive
Technologies I work with: Kafka, Flink, Spark, Airflow, dbt, Iceberg, Redshift, BigQuery, Kubernetes, ArgoCD, Terraform, Docker, Ray, Hive, OpenSearch, Grafana, Prometheus
01 Architecture
Systems I built
Conceptual. Client topology not reproduced.
02 Record
Ten years
- 2025 – nowNCubeLead & solutions architect / Logistics
- 2021 – 2025ExadelLead & solutions architect / Healthcare, automotive
- 2019 – 2021SoftServeMLOps & data engineer / Banking
- 2018 – 2019XperraMachine learning engineer / NLP, vision / Founded by a SoftServe board director
- 2017 – 2018ProsapientData engineer / Analytics
- 2017DataRobotData analyst / Automated ML
03 Public
In the open
PageIndex: success does not mean every requested page came back
Paperclip’s $0 budget is not a stop button
Hindsight’s agent memory: how four second places become first
The classifier’s 100% came from two missing labels
The Ember-1 headline is 40%. The useful number is a range.
A NULL receipt hid two pending jobs
Investment Products Recommendation Engine
MLOps: why should I care?
04 Contact
Open a channel
anton.vlacenko@gmail.com LinkedIn GitHub X Medium CV
© 2026 Anton Vlasenko End of transmission