Work

Two decades of one commit history: physics → big data → ML → GenAI. Also available as a PDF.

1. genai · agents
2. Sep 2022 — present · Remote

   Provectus

   Solution Architect / Senior Data Engineer
   • Lead design and deployment of production-grade GenAI systems — multi-agent workflows (LangChain, CrewAI, AWS Bedrock) for enterprise applications.
   • Built large-scale RAG pipelines processing 12M+ medical records/day, integrating OCR, classification, and retrieval for clinical data access.
   • Designed LLM-powered analytics for performance tracking and cost optimization — cut infrastructure costs by 45%.
   • Developed Model Context Protocol (MCP) servers with auth, streaming, and structured tool interfaces for text-to-SQL flows.
   • Orchestrated genomics (Omics) pipelines on AWS with Airflow — DynamoDB metastore, SQS/SNS queues, Lambda triggers, heavy batching.
   pythonbedrocklangchaincrewaiairflowhealthomicsopensearchdynamodb
3. big data · ml
4. May 2020 — Sep 2022 · Remote

   Kinner Inc

   Data Engineer / Data Scientist (contract)
   • ETL development in Python/R with Airflow across MySQL, NoSQL, Spark, and Hadoop in ad-tech.
   • Reporting dashboards and alerting wired into Tableau and Slack.
   • Partner KPI analytics, fraud detection, user retention, and ROI forecasting.
   pythonrairflowsparkhadoopmysqltableau
5. Jan 2020 — May 2020 · Montreal

   Gazelle AI

   Machine Learning Engineer
   • Built text preprocessing frameworks for supervised NAICS-code classification with BERT-based NLP models.
   • Ported and optimized proprietary MATLAB classification algorithms to Python.
   • Data lake and pipeline development with AWS Athena/Glue; migrated services from MongoDB to DynamoDB and introduced Airflow.
   pytorchbertdaskglueathenamongodb
6. Dec 2018 — May 2020 · Montreal

   Concordia University

   Course Instructor, part-time
   • Developed and taught CEBD 1160 "Introduction to Big Data Technology" — Hadoop, Spark, data wrangling in R and Python, workflow automation, databases, and cloud.
   • Led the restructuring of the course into separate modules.
   hadoopsparkrpython
7. Mar 2018 — Aug 2018 · Montreal

   SSENSE

   Senior Data Engineer
   • Batch ETL pipelines in Apache Airflow on AWS EMR (Spark, Hive), S3, and Redshift.
   • Built the GDPR pipeline service anonymizing and removing user data across S3 buckets and multiple databases.
   • Integrated AWS Glue/Athena to replace the legacy ETL pipeline.
   pythonairflowemrsparkredshiftathena
8. Oct 2017 — Mar 2018 · Montreal

   Bandsintown

   Back-End Developer
   • Tuned the Hadoop cluster; added and configured Spark and Hive-on-Spark, hardened cluster security.
   • Wrote custom MapReduce jobs integrated with the pipeline (Java, Oozie).
   • R&D for AWS Glue and Athena adoption with the existing infrastructure.
   hadoopsparkhivejavaoozie
9. Oct 2015 — Sep 2017 · Montreal

   Wajam

   Big Data & BI Developer
   • Built data collection, processing, and visualization pipelines for current and new products (Scribe, Scala, Pig, Hive, Sqoop, Tableau).
   • Administered a 60-node Cloudera Hadoop cluster ingesting 100+ GB of raw data hourly.
   • Developed business KPI forecasting and Slack-integrated alerting; optimized Hive with custom Java UDFs/UDAFs.
   hadoopscalapighivemysqltableau
10. physics
11. 2010 — 2017 · Albany, NY

    University at Albany, SUNY

    PhD Physics (All But Defence) · M.S. Physics

    Dissertation: stochastic Liouville equation modeling for NMR line shape calculation (SpinAl software) + supervised machine learning on Raman spectroscopy data. Coursework in computational methods, quantum mechanics, ML, Bayesian inference.

12. 2006 — 2010 · Kharkiv, Ukraine

    V.N. Karazin Kharkiv National University

    BSc Physics

    Theoretical physics, discrete mathematics, linear algebra, algorithms, distributed computing. Where the track begins.