Anna is an experienced AI/ML Engineer with a strong background in MLOps, Generative AI, and DevOps. She has successfully delivered production-level code and automated CI/CD pipelines for ML model deployment. With expertise in both Azure and AWS, Anna optimizes processes and drives innovation across various industries including BFSI, Healthcare, and SaaS. Her technical proficiency includes Python, Azure ML, and OpenAI, enabling her to develop impactful AI solutions.
Successfully delivered production-level code by translating data scientist needs.
Implemented automated CI/CD pipelines for ML model deployment and monitoring across multiple projects.
Spearheaded the automation of machine learning workflows for scalability and reliability.
Increased coder efficiency by 30% by streamlining debugging and automating code generation.
Reduced diagnostic review times by 25% and enhanced prediction accuracy in Radiology AI for bone fracture detection.
Achieved 95% accuracy in identifying at-risk customers and reduced churn by 20%.
Overview: Developed a web application leveraging GPT-4 in Azure OpenAI for code correction and generation. Responsibilities: Designed and implemented the web application on Azure. Integrated GPT-4 via Azure OpenAI for core AI functionalities. Utilized RAG for data customization and Assistant API for structured outputs. Deployed the solution via Azure Web App.
Key outcomes:
Increased coder efficiency by 30%.
Overview: Developed a computer vision-based API to predict bone fractures using YOLOv8, Faster R-CNN, and U-NET. Responsibilities: Developed a computer vision-based API leveraging YOLOv8, Faster R-CNN for classification and detection. Implemented MLOps processes including automated CI/CD pipeline with Azure DevOps and Azure ML.
Key outcomes:
Reduced diagnostic review times by 25%.
Overview: Developed a churn prediction system using Python, Azure ML, and Power BI. Responsibilities: Developed the churn prediction system using Python and deployed it on Azure ML. Utilized Power BI to analyze customer data and visualize insights.
Key outcomes:
Achieved 95% accuracy in identifying at-risk customers.
Recommendation System for Prospect Identification — ML-based system for identifying similar + high-potential prospects. 40% success-to-approach lift.
Key outcomes:
Improved the success-to-approach ratio by 40%, enabling more targeted marketing.
Increased the overall efficiency of customer acquisition and engagement strategies.
Supply Chain Optimization — ML model for inventory cost reduction + delivery efficiency in manufacturing + pharma.
Key outcomes:
Reduced inventory costs and improved delivery efficiency.
Anna
MLops Developer