Raya Kichukova
Equally at home tracking plant roots with RL and deploying tools that teach Dutch sign language.
Data science student who has shipped for real clients — IKEA dashboards, production deployments, and systems that solve problems people actually have. I like work with stakes.
Projects
A selection of things I have built — from research pipelines to client deployments.
Autonomous Root Tip Inoculation System
A fully autonomous pipeline combining deep semantic segmentation, graph theory tracing, and PID/RL controllers to replace manual pipetting in high-throughput plant phenotyping at NPEC.
SDG Internet & Education Dashboard
A Power BI dashboard analysing the correlation between internet access and educational completion rates across South America from 2000 to 2023, aligned with UN SDG Goals 4 and 9.
SignSee — Sign Language Recognition MLOps Application
A production-grade MLOps application for real-time NGT sign language recognition, deployed across local, on-premises, and Azure environments with full CI/CD and operational monitoring.
Skills
From raw data to deployed systems — spanning computer vision, machine learning, NLP, and full-stack development with a track record of shipping for real clients.
Computer Vision
Image segmentation, object detection, semantic masks, root tracing with graph algorithms, model evaluation on real-world data. Experience with U-Net, YOLO, and custom CV pipelines.
Machine Learning & Deep Learning
Supervised and unsupervised learning, model training and evaluation, PyTorch and scikit-learn. Experience with medical data, time series, and production ML pipelines on Azure ML.
NLP & RAG
Emotion prediction from unscripted text in Bulgarian, RAG pipeline development for return system optimisation, language model integration and prompt engineering.
Reinforcement Learning & Control
Soft Actor-Critic (SAC) agents, PID controllers, MuJoCo simulation environments, reward function design, and sim-to-real coordinate transformation pipelines.
Data Analysis & Visualisation
Python data pipelines with pandas and NumPy, Power BI dashboards for IKEA and SDG research, statistical analysis, correlation and regression modelling.
Deployment & MLOps
Azure Container Apps, Azure ML, Docker, Portainer, GitHub Actions CI/CD, MLflow experiment tracking, blue/green model deployments, and on-premises server management.
Full-Stack Development
FastAPI backends with WebSocket support, Vue 3/TypeScript frontends, PostgreSQL with async SQLAlchemy, REST APIs, and operational monitoring dashboards with Chart.js.
Collaboration & Delivery
Agile delivery in university and client projects, Azure DevOps sprint management, peer code reviews, and direct client work with IKEA and a logistics company.
Education
Breda University of Applied Sciences
2024 — 2028 (expected)B.Sc. Applied Data Science and Artificial Intelligence
Covering machine learning, computer vision, NLP, reinforcement learning, data analysis, and full-stack development. Worked on client projects for IKEA and a logistics company alongside academic research.
About Me
I am a third-year Applied Data Science & AI student at Breda University of Applied Sciences with a growing focus on computer vision and robotics. I care about making my work not just functional but explainable and well-presented — because a result nobody can interpret is only half a result. I have a tendency to notice the things others skip over, which makes the difference between something that works and something that works reliably.
Quick Facts
Education
B.Sc. Applied Data Science & AI, Breda University of Applied Sciences (2024 — 2028)
Location
Breda, The Netherlands
Interests
Languages
Let's Connect
Contact Information
Let's connect! I'm always interested in hearing about new projects and opportunities.