Daniil Kurilo Email & Phone Number

Minsk, BY

Poland

• Skin diseases detection engine (CV):
• Proposed and implemented new solutions that improved model accuracy by 10%.
• Designed and developed multi-stage training pipelines utilizing AWS training jobs.
• Configured infrastructure for seamless data and weight uploads, as well as checkpoint exports to S3 Bucket.
• Performed extensive data cleaning and annotation to ensure high-quality training datasets.

Poland

• Chatbot for sales staff (NLP):
• Experimented with different LLMs for various tasks (NER, summarization, key phrase extraction, etc.).
• Set up a vector database using ChromaDB for storing client requests and employee information.
• Experimented with various embedding models for vectorizing custom domain specific data.
• Created a preprocessing pipeline that extracts useful insights from the chat with the client and saves in databases.
• Created custom Langchain agents to complete different tasks based on the bot functionality(q&a, recommendation query formulation, entity extraction, etc.).

• Over the course of seven intensive month, I immersed myself in the study of a comprehensive machine learning program. This course covered a large number of different topics and practical applications.Сlassical Machine.
• Delved into supervised learning algorithms (linear regression, logistic regression, support vector machines, decision trees, ensemble methods).
• Delved into feaeture engineering, regularization techniques, hyperparameter optimization.
• Delved into unsupervised learning algorithms (k-means clustering, hierarchical clustering, DBSCAN, PCA).Deep Learning and Neural Networks:
• Delved into fundamentals of deep learning | neural network architectures and their applications in Computer Vision | Natural Language Processing | Time Series analysis.
• Utilized PyTorch for deep learning tasks, implementing forward and backward error propagation, gradient descent, and optimization methods.

Bachelor'S Degree, Computer Software Engineering

Data Science