← Home

Download PDF

PDF · 222 KB

I design and ship end-to-end machine learning and LLM systems, from quick prototypes to high-load production. My background is in distributed systems, and lately most of my work is LLMOps and agentic workflows. I care about architectures that stay maintainable and about building what the product actually needs.

Renesas Electronics

Nov. 2026 – Present
Senior Staff AI AnalystTokyo, Japan
  • Part of the AI Strategy, Governance & Enablement team in the Information Systems Division of a global semiconductor company.
  • Design, build and deploy AI-powered tools that bring LLM APIs and ML models into products and internal workflows.
  • Work with business teams to turn their needs into AI-driven features, from building the solution to supporting it in use.

GrainFox

Feb. 2023 – Oct. 2026
AI Systems EngineerWinnipeg, Canada (Contract)
  • Architected and deployed a production-grade Agentic Ecosystem of state-driven, multimodal agents running diverse business workflows end-to-end with an 85% autonomous task completion rate.
  • Engineered a Generative Analytics engine (RAG) over a continuously updated knowledge base (LlamaIndex, hybrid vector search, optimized chunking), cutting manual research time by 90%.
  • Built LLMOps infrastructure on GCP Cloud Run with a custom gateway for prompt management and smart model routing: full provider independence and a 30% cut in inference costs.
  • Shipped human-in-the-loop tooling (React, TypeScript, MCP) that lets operations teams supervise and correct agent decisions without engineering involvement.

Yandex

Feb. 2024 – Jan. 2025
Machine Learning Engineer, Search AntifraudMoscow, Russia
  • Engineered high-throughput SERP antifraud models (CatBoost + BERT embeddings) served via low-latency Python APIs (<20ms) across millions of requests daily.
  • Mined petabyte-scale clickstream logs (MapReduce, SQL/YQL) for adversarial behavioral patterns — lifted fraud detection precision by 6% without sacrificing recall.
  • Owned reliability of distributed model serving end-to-end, cutting production incident response time by 40% for search clusters.

Inkast

Feb. 2022 – Jul. 2023
Machine Learning EngineerSt. Petersburg, Russia
  • Developed a hybrid GNN-based Recommender System behind a Django API — 12% uplift in user retention from better suggestion relevance.
  • Contributed across the stack: PostgreSQL schema design and Vue.js visualization components delivering model insights directly to end users.

Education

St. Petersburg Polytechnic UniversitySt. Petersburg, Russia
  • M.Sc. in Biomedical Computing2021 – 2023
  • B.Sc. in Medical Biophysics2017 – 2021

Publications

  • Development of CNN for Defining a Renal Pathology using CT ImagesSpringer (2022)
  • Evaluation of Haptoglobin and Its Proteoforms as Glioblastoma MarkersIJMS (2021)

Technical Skills

GenAI & LLMOps

Agent Frameworks LangChain, LangGraph, AutoGen
RAG Pipelines LlamaIndex, Haystack, GraphRAG
Evaluation Ragas, DSPy, LangSmith
Vector Infrastructure Vertex AI, Pinecone, Weaviate

Cloud & MLOps

Cloud Platforms GCP, AWS
Infrastructure as Code Terraform
Model Serving vLLM, Triton
Containerization Kubernetes, Docker
Observability Prometheus, Grafana

Data & ML Engineering

Distributed Computing Spark, MapReduce, SQL/YQL
Workflow Orchestration Airflow, MLflow
Deep Learning PyTorch, TensorFlow
Classical ML CatBoost

Software Architecture

Backend Python, FastAPI, Django
Frontend Vue.js, React, TypeScript
Data Stores PostgreSQL, Redis, MongoDB
API Standards REST, OpenAPI, gRPC

Technical Skills

GenAI & LLMOps

Agent Frameworks
LangChain, LangGraph, AutoGen
RAG Pipelines
LlamaIndex, Haystack, GraphRAG
Evaluation
Ragas, DSPy, LangSmith
Vector Infrastructure
Vertex AI, Pinecone, Weaviate

Cloud & MLOps

Cloud Platforms
GCP, AWS
Infrastructure as Code
Terraform
Model Serving
vLLM, Triton
Containerization
Kubernetes, Docker
Observability
Prometheus, Grafana

Data & ML Engineering

Distributed Computing
Spark, MapReduce, SQL/YQL
Workflow Orchestration
Airflow, MLflow
Deep Learning
PyTorch, TensorFlow
Classical ML
CatBoost

Software Architecture

Backend
Python, FastAPI, Django
Frontend
Vue.js, React, TypeScript
Data Stores
PostgreSQL, Redis, MongoDB
API Standards
REST, OpenAPI, gRPC