Alona Liuzniak-AI Architect

AI-powered platform for automated UX validation and designer support

• Designed and led technical implementation of an enterprise-wide AI solution for automated UX review that improved design quality and significantly reduced manual review processes in teams
• Developed an automated UX validation tool as a Figma plugin and web application that generates test cases based on internal guidelines and reliably checks current designs for consistency and standard compliance
• Implemented an interactive designer chat based on RAG that answers questions about the current design and the company's UX guidelines, and designed the deployment architecture using containerized services
• Python, Azure OpenAI, PostgreSQL, REST API, Docker, OpenShift, Helm, CI/CD, Figma MCP, LLM, RAG, Prompt Engineering, GenAI, XAI, AI Architecture, AI Strategy

Chatbot solution for customer support with RAG in the chemical industry

• Led the design and implementation of an end-to-end chatbot solution for customer support and product recommendations, delivering a fully functional system within three months
• Improved search result quality in the chatbot knowledge base by using prompt engineering, summarization techniques, and a two-stage search system based on product name filtering and matching product relationships to handle complex queries
• Developed a modular chatbot architecture to create reusable and customizable pipelines and designed the deployment architecture using containerized services
• Integrated responsible AI mechanisms that provide users with transparent and understandable explanations for product recommendations
• Python, Microsoft Azure, MLflow, Docker, CI, Prompt Engineering, LLM, RAG, NLP, GenAI, XAI (Explainable AI), RAI (Responsible AI), AI Architecture, AI Strategy

AI Strategy & AI Adoption

• Designed and conducted workshops to demonstrate generative AI potentials, risks, and architectural considerations for successful application in business units
• Delivered a webinar on building production-ready GenAI systems considering regulatory frameworks, hallucination and bias reduction, and real use cases in the financial sector
• Translated complex AI topics into practical architectural guidelines for product owners, developers, and management to increase organizational readiness for AI adoption
• Acted as liaison between legal, business, and technical teams by translating regulatory and domain requirements into actionable AI architectures
• AI Architecture, AI Strategy, AI Governance, Risk Analysis, Hallucination Avoidance, Bias Reduction, Prompt Design, RAG Architecture Patterns, AI Adoption Strategies

Chatbot solution with image integration and communication flow optimization

• Led the design and implementation of an end-to-end chatbot solution for customer support and product recommendations, delivering a fully functional system within three months
• Improved search result quality in the chatbot knowledge base by using prompt engineering, summarization techniques, and a two-stage search system based on product name filtering and matching product relationships to handle complex queries
• Implemented a modular architecture to build customizable chatbot pipelines
• Integrated responsible AI mechanisms that provide users with transparent and understandable explanations for product recommendations
• Python, Microsoft Azure, MLflow, Docker, CI, Prompt Engineering, LLM, RAG, NLP, GenAI, XAI (Explainable AI), RAI (Responsible AI)

Solution for fraud prevention in the banking sector

• Significantly simplified model training, deployment, and evaluation by introducing an effective customer segmentation strategy that reduced the number of fraud prevention models from 10 million to 9 models
• Optimized the existing XGBoost model by developing new features
• Implemented a new transaction assessment strategy based on an autoencoder model
• Python, Java, Pandas, NumPy, Scikit-Network, Pyplot, JUnit, FitNesse, HDFS, Apache HBase, Spring Boot, Apache Kafka, PySpark, Docker, Machine Learning, XGBoost, Graphs, Clustering, AFC, Fraud Prevention, Jira, Confluence

• Developed an object detection solution in Python that performed real-time inference and enabled efficient image capture and decision-making in production environments
• Generated synthetic training data for machine learning models to simulate object detection and handling tasks
• Applied reinforcement learning techniques to optimize robot motion planning
• Python, TensorFlow, NumPy, OpenCV, Pytesseract, Pandas, Scikit-Learn, Pytest, Blender, Deep Learning, Computer Vision, Jira, Confluence