Responsibilities

• Core contributor to the design and development of the company’s flagship AI testing and auditing platform, collaborating with AI engineers, data scientists, and software engineers.
• Led feasibility analysis and implemented 5+ critical AI features, including model performance evaluation, fairness/bias testing, explainability analysis, statistical analysis and drift detection.
• Converted 2+ successful research prototypes into production ready features, improving execution efficiency and computational performance.
  
  

Key Achievements

• Accelerated platform readiness for MVP and alpha releases by designing and implementing core AI evaluation features.
• Delivered live technical demos and presentations to 10+ enterprise clients, contributing to a 80% customer retention rate.
• Automated 60% of the platform’s core evaluation workflows, cutting manual assessment efforts in half and improving efficiency.

Responsibilities

• Collaborated with clients to identify business challenges and opportunities where AI technologies could be applied effectively.
• Conducted AI system audits and assessments for 10+ enterprise AI models to identify issues under tight deadlines:
  • Evaluated the performance, fairness, security and explainability of AI systems and helped the client’s engineers to design and implement custom AI solutions tailored to their needs, utilizing ML algorithms, DL models, and NLP techniques.
  • Conducted comprehensive data analysis and interpretation to derive actionable insights and recommendations for optimizing clients' processes and decision-making.
• Provided technical guidance on AI model tuning, data preprocessing, and optimization to cross- functional teams, ensuring successful delivery within scope, timeline, and budget constraints.
• Delivered client presentations and training sessions to educate stakeholders on AI concepts, project progress, and outcomes.
  
  

Key Achievements

• Contributed to AI due diligence efforts for 2 major acquisitions, evaluating technical feasibility, scalability, and potential risks in AI models and data infrastructure, contributing to successful investment decisions.
• Conducted comprehensive analysis of 3 AI systems for a leading bank, identifying critical issues related to model performance, bias/fairness, and compliance with regulatory standards (e.g. GDPR, EU AI Act)
• Developed an AI risk assessment tool, reducing AI compliance validation time by 50%.

Responsibilities:

• Performed data annotation, preparation and engineering for training AI models.
• Data visualization with Plotly, Matplotlib and Seaborn libraries.
• Applied Machine Learning unsupervised (Self-Organizing Maps, Gaussian Mixture Models) and supervised learning (Random Forest, SVM, XGBoost, AdaBoost, Logistic Regression, KNN, CART) algorithms to real-world AI problems.
• Facilitated discussions with project partners and other stakeholders.
  
  

Key Achievements

• Involved in 2 EU-Funded (Horizon 2020 research and innovation program) projects (Kyklos 4.0 and Stargate) developing AI solutions for sustainability and manufacturing.
• Prepared part of a technical deliverable for Task 3.5 of Kyklos 4.0 Project, influencing EU-funded AI research initiatives.
• Built a deep learning model (U-Net) in Keras Tensorflow for a semantic image segmentation task (weed detection) achieving 92% accuracy on test data, reducing manual weed removal efforts by 40% through automated identification.