Dmytro — Data Scientist

Work Experience

Data Scientist (AI-powered dynamic pricing and demand forecasting system)

Duration: 6 months

Summary: Developed an AI-powered dynamic pricing and demand forecasting system that optimizes weekly tour prices by analyzing booking history, product data, and operational constraints to forecast demand, estimate price elasticity, and recommend revenue-maximizing prices.

Responsibilities:

• Led the AI part of the project, including new feature development and maintenance/bug fixing.
• Performed model research and selection for upcoming functionality.
• Drove the new Bareboats module, worked with new data types and defined the ML approach.
• Maintained and updated the existing system, including implementing daily update pipelines.
• Ran testing and research based on client requests, covering both ML tasks and domain questions.

Technologies: Python, Pandas, NumPy, Statsmodels (SARIMA), Prophet, XGBoost, SciPy, SLSQP

Data Scientist (WDC Platform)

Duration: 2 years

Summary:

• Built a system for clustering and summarizing news texts using embeddings, deep learning, and topic modeling techniques
• The pipeline combined OpenAI/Nomic embeddings, autoencoder-based dimensionality reduction, and HDBSCAN clustering to analyze and summarize news content efficiently

Responsibilities:

• Worked on text embedding, classification, and summarization tasks.
• Trained, tested, and tuned classical ML models and neural networks using PyTorch.
• Conducted research-focused tasks, evaluating various approaches and optimizing solutions for limited hardware resources.
• Developed and deployed small RESTful API services for integration into the overall product.
• Collaborated in a small team to maintain and enhance legacy text analysis services daily.

Technologies: Python, PyTorch, OpenAI API, Tiktoken, Langchain, Hdbscan, Docker, Nvidia container toolkit

Data Scientist (Stock Piper Platform)

Duration: 6 months

Summary: Developed a web platform that automatically captures news and press releases related to copper stocks in the commodities market, transforms this information using AI into professional journalistic articles, and publishes them automatically.

Responsibilities:

• Created a processing script for Amazon Textract to collect and filter input text.
• Built a post-processing module and implemented OpenAI model for text generation.
• Enhanced performance using token optimization with Tiktoken.
• Applied prompt engineering techniques to generate and summarize news content.

Technologies: Python, OpenAI API, AWS (DynamoDB, Textract, S3), Tiktoken, Pandas

Data Scientist (Omnitrend Voice AI)

Duration: 6 months

Summary: Developed an AI-powered conversational assistant designed to conduct pre-election political surveys in Australia, introducing users to candidates, explaining their positions, and collecting user opinions on each candidate’s views.

Responsibilities:

• Designed and refined the conversational flow for smooth interactions.
• Built and configured the chatbot end to end.
• Optimized bot settings for realistic voice and accurate data extraction.
• Implemented fallback logic for unexpected user inputs.
• Ran demos and performed regular testing to ensure quality.

Technologies: Python, Dialogflow ES, Twilio, NLTK, spaCy

Data Scientist (Virtual Agent for TechSupport System)

Duration: 6 months

Summary: Created a virtual agent for FAQs and ticket logging on a TechSupport system, capable of extracting user information and issue details during conversations to fill tickets or provide self-help guidance.

Responsibilities:

• Designed conversational flow and developed chatbot for realistic conversation.
• Created a system for managing connection to TechSupport system for ticket creation.
• Extracted detailed information from conversations and post-processed conversational records and text.
• Implemented issues classification system for smarter ticket assignment.
• Integrated OpenAI Whisper for improving sound processing.

Technologies: Python, Dialogflow CX, Spacy, NLTK, Twillio, Google Firestore, OpenAI API, GCP

Education

• NTUU "KPI", Institute for Applied System Analysis
  Master of System Analysis
  Master