I design, build, and deploy production-grade ML systems — from data ingestion and feature pipelines to model serving, monitoring, and business metrics.
Analytics → Machine Learning → Production ML / MLOps
- End-to-end ML systems in fintech & industrial domains
- Credit risk modeling (PD / LGD), regulatory ML
- Time series & NLP systems with business impact
- Strong emphasis on reproducibility, monitoring, and validation
ML Engineer — Industrial / Mining (R&D)
- Anomaly detection systems for industrial data
- OCR models and data processing pipelines
- Full ML lifecycle: dataset validation → training → deployment → monitoring
- MLOps stack: Airflow, DVC, ClearML, S3-compatible storage
ML Engineer — Fintech (Banking)
- Credit risk models for corporate clients
- Models developed under regulatory requirements
- Production deployment via Kubeflow & CI/CD pipelines
- Oracle-based data pipelines and automated data collection
ML Engineer — Product / Startup
- Built a full ML product from scratch (ETL → models → backend)
- Multisource ingestion: blockchain, exchanges, social & news data
- NLP sentiment indices (FinBERT) and Prophet-based forecasting
- Dockerized microservices and production-ready architecture
🚀 Production ML Pipelines
- Python, SQL, Airflow
- Data ingestion → feature engineering → model training → metrics
- Dataset validation, versioning (DVC), experiment tracking
- Used in real production environments
📈 Credit Risk & Regulatory ML
- PD / LGD models for corporate lending
- Model validation & business-oriented metrics
- Work under banking regulations (model governance, reproducibility)
- Close interaction with data, risk, and business teams
🧠 NLP & Time Series Systems
- FinBERT / BERT for news & social sentiment
- Multivariate & multitype time series forecasting (Prophet)
- Sentiment indices, dynamic regressors, feature attribution
🤖 ML Services & Automation
- FastAPI-based ML backends
- Telegram bots for analytics, signals, and alerts
- Docker & docker-compose deployment
- Python
- pandas, NumPy, SciPy
- scikit-learn
- PyTorch
- XGBoost, CatBoost, LightGBM
- Prophet (multivariate & dynamic regressors)
- PostgreSQL, MySQL, Oracle, ClickHouse
- SQLAlchemy
- ETL / ELT pipelines
- Feature engineering & dataset validation
- Hadoop ecosystem (basics)
- AWS S3
- Yandex Object Storage (S3-compatible)
- Blockchain & crypto APIs
- External data sources (REST / GraphQL)
- Docker, Docker Compose
- Linux
- Git, GitLab
- CI/CD pipelines
- DVC (data & model versioning)
- Experiment tracking: MLflow, ClearML
- Apache Airflow
- Kubeflow
- FastAPI
- REST APIs
- Async Python (asyncio, aiohttp)
- ML microservices
-
🥇 Winner — International Student Olympiad “Engineering Breakthrough” (2025)
(Track: Data Engineering & Machine Learning) -
🥉 Bronze Medalist — All-Russian Olympiad “I Am a Professional” (2025)
(Business Informatics) -
🎯 Finalist — Sovcombank Case Championship (2024)
(Credit risk modeling) -
🎓 Assistant Lecturer in Machine Learning (2024-2025)
Teaching seminars, reviewing labs, and mentoring students
