Contact information

You can get this candidate's contact information from https://www.work.ua/resumes/20020738/

Work experience

AI Software Engineer

from 02.2023 to 07.2026 (3 years 6 months)
Modal, Remote (IT)

• Built serverless GPU inference services and AI model deployment infrastructure across Modal's multi-cloud platform spanning AWS, Google Cloud, and Oracle Cloud Infrastructure (OCI).
• Developed distributed batch processing pipelines for large-scale inference, embedding generation, and AI workloads using Python, PyTorch, and containerized execution environments.
• Reduced cold-start latency for GPU-backed inference services by optimizing container initialization, model loading, image caching, and worker warm-up to improve request responsiveness.
• Designed GPU-aware autoscaling and request scheduling strategies to maintain low-latency inference while maximizing GPU utilization during fluctuating production traffic.
• Built production APIs and asynchronous execution workflows for deploying transformer models, supporting high-throughput inference and reliable request processing.
• Optimized distributed batch execution by implementing workload partitioning, retry mechanisms, and fault-tolerant processing for long-running AI jobs across multiple compute workers.
• Implemented observability, performance profiling, and operational monitoring to identify latency bottlenecks, GPU resource utilization, autoscaling behavior, and production service reliability.

Senior Software Engineer

from 06.2021 to 02.2023 (1 year 9 months)
Weaviate, Remote (IT)

• Built RAG workflows integrating document loaders, text chunking, embeddings, and vector stores such as FAISS and Pinecone for semantic retrieval and question-answering applications.
• Developed Generative Search and RAG integrations by connecting document ingestion, embeddings, metadata filtering, and vector search to improve context retrieval for LLM-powered applications.
• Improved hybrid search quality by balancing semantic vector similarity with BM25 keyword ranking, tuning retrieval logic and score normalization to deliver more relevant search results.
• Designed cloud-native APIs and distributed services using Python, GraphQL, Docker, and Kubernetes to support scalable vector search and retrieval workloads.
• Optimized indexing and query performance for large vector collections through efficient HNSW indexing, metadata filtering, and asynchronous data ingestion pipelines.

Software Engineer

from 06.2020 to 06.2021 (1 year)
Agility Robotics, Remote (IT)

• Built Python backend services for Digit robot telemetry, diagnostics, and fleet remote-monitoring workflows on AWS-based infrastructure.
• Developed real-time ingestion pipelines for robot health, sensor, diagnostic, and task-status events using Kafka, Redis, and PostgreSQL.
• Solved intermittent robot connectivity issues by implementing local event buffering, sequence-based synchronization, retry logic, and deduplication to recover telemetry safely after reconnection.
• Containerized telemetry and monitoring services with Docker and Kubernetes, improving service scalability and reliability for distributed robot fleet operations.

Backend Engineer

from 04.2018 to 06.2020 (2 years 3 months)
Genpact, Remote (IT)

• Built backend services for LazMall and marketplace seller/order workflows on Alibaba Cloud–based infrastructure, developing APIs for product, SKU, inventory, and order processing.
• Solved inventory overselling during high-traffic flash sales by implementing atomic stock updates, inventory reservation logic, and concurrency controls to prevent multiple orders from consuming the same SKU availability.
• Developed event-driven order and inventory workflows using Kafka, Redis, and MySQL, improving backend scalability and consistency during major campaign traffic spikes.

Python developer

from 04.2015 to 04.2018 (3 years)
Celonis Munich, Remote (IT)

• Built Python and Django backend services for GCash bills payment and prepaid load workflows, integrating wallet transactions with biller and telecom partner APIs through REST services and asynchronous processing.
• Designed idempotent transaction handling with unique transaction references, state validation, and controlled retries to prevent duplicate payments when customers retried after network or partner timeouts.
• Contributed to Scan-to-Pay QR backend services by validating merchant IDs, timestamps, and signed payment payloads to reject invalid, expired, or manipulated requests before transaction processing.

Education

Odesa Polytechnic National University

Computer Science, Odesa
Higher, from 2011 to 2015 (4 years)

Bachelor of Science in Computer Science, with a strong foundation in software development, programming, databases, algorithms, data structures, and computer systems. Completed the degree with a final grade of 9.30/10 between October 2011 and September 2015.

Additional education and certificates

No additional education or certificates.

Knowledge and skills

• Python
• SQL
• PySpark
• C#
• Java
• C/C++
• JavaScript
• Apache Spark
• Kafka
• Spark Streaming
• Airflow
• Dbt
• Feature Engineering
• ML Pipelines
• Generative AI
• RAG
• Vector Da
• PostgreSQL
• MySQL
• SQL Server
• Pgvector
• ETL/ELT
• Data Pipelines
• CDC
• Data Modeling
• Data Quality
• Data
• AWS
• Azure
• GCP
• Databricks
• Snowflake
• BigQuery
• Delta Lake
• Docker
• Kubernetes
• CI/CD
• Monitoring
• Cloud Deployment
• .NET Core
• ETL
• Productivity
• Git
• Conflict-free
• HTML
• CSS
• Manual testing
• Knowledge of OOP principles
• Writing bug reports
• Jira

Language proficiencies

English — advanced