Public profile
@detrix
Backend/ML Engineer
Graduated from Nazarbayev University with a Bachelor's and a Master's Degree in Computer Science, with experience in research and coding.
Experience, skills, projects, and achievements.
Backend/ML Engineer
Analyzed tens of millions of onboarding events and benchmarked CatBoost, Random Forest, and XGBoost models; identified KYC verification as the primary conversion bottleneck and proposed product improvements to reduce customer drop-off
Built a FastAPI and PostgreSQL service connecting an internal merchandise store with a Bitrix24-based supplier, synchronizing catalog and stock data and managing order reservation, cancellation, and status callbacks; implemented HMAC signing and API key authentication, idempotency, Alembic migrations, scheduled synchronization, webhook auditing, and automated tests with pytest
Built an ETL and analytics platform combining amoCRM, clinic ERP, and marketing spend data through entity matching and deduplication; implemented webhook and polling synchronization, PostgreSQL storage, and FastAPI endpoints for ROMI, CAC, LTV, funnel, retention, cohort, and revenue metrics
Nazarbayev University
May 2025 — PresentForecasting Pipeline: Built a weakly supervised time-series pipeline for four wheat diseases, processing 666,000+ hourly records per location and engineering rolling weather features over 3–72-hour windows; Model Evaluation: Benchmarked XGBoost, Random Forest, Logistic Regression, 1D CNN, LSTM, and GRU models, and evaluated generalization across three climatic regions;
italita.kz
Mar 2025 — Dec 2025Backend Platform: Built an asynchronous FastAPI service integrating amoCRM, iiko, Yandex Delivery API, and PostgreSQL for multi-location restaurant order processing and analytics; Performance & Testing: Reduced order ingestion latency from 3.2 s to 1.3 s (59%) and tested REST endpoints and integration workflows using pytest and Postman;
Nazarbayev University
Feb 2024 — Dec 2024Reinforcement Learning: Built a custom Gymnasium/SimPy environment for a Time-Slotted LoRa simulator and trained a DQN agent to select 1–3 uplink attempts from PRR, RSSI, and spreading-factor observations; Optimization & Engineering: Reduced uplink transmissions by approximately 47% while maintaining 0.90–1.00 PRR and refactored a monolithic 1,000+ line codebase into modular components;
Aspan Technologies
Jul 2023 — Mar 2024Rover Software: Developed Python and C++ ROS components integrating GPS, cameras, motor control, and remote operator commands for a rover competing in the University Rover Challenge 2024; Communication Systems: Integrated LoRa rover-to-base communication and prototyped WebRTC camera streaming with WebSocket signaling, OpenCV, and fail-safe connection handling;
Nazarbayev University
2020 — 2024cGPA: 3.17/4.00 (Dean's List Spring 2024)
Nazarbayev University
2024 — 2026cGPA: 3.42/4.00 (Ranked 7th/37)
A compact preview of this member's public research layer.
Working as a Research Assistant at Nazarbayev University.
