Asma Irfan

• Engineered end-to-end ML pipelines combining Random Forest and supervised ANN models in TensorFlow and Pandas to meteorologically adjust PM2.5 concentrations across 5 Indian monitoring sites, measurably improving forecast accuracy. • Developed and deployed production ML models in TensorFlow using transfer learning, advanced feature engineering, and cross-validation; tracked experiments with MLflow to ensure reproducibility and model reliability. • Exposed trained models as REST APIs using FastAPI and Docker, enabling real-time inference and integrating model serving into production data workflows via Azure ML. • Designed interactive Power BI and Tableau dashboards to surface model outputs and air quality insights, enabling faster, evidence-based decision-making for scientists and business stakeholders. • Worked with scientists, engineers, and executives to design AI solutions that met project requirements, were deployed, and were monitored, which reduced development time and increased stakeholder satisfaction

Assisted in designing and delivering course materials, including lectures, assignments, and exams for Scientific Data Visualization graduate students. • Managed and graded assignments and exams for 250+ graduate students, providing constructive feedback to enhance analytical skills

• Reduced deployment cycle times by 40% by engineering and automating CI/CD pipelines using Azure DevOps, accelerating release velocity across engineering teams. • Cut incident response time by 30% by configuring monitoring and drift-alerting infrastructure with Prometheus, Grafana, and Checkmk across 100+ CentOS/Ubuntu servers. • Managed Azure cloud resources across 3 of 9 critical infrastructure domains, improving system reliability and uptime for business-critical applications. • Automated infrastructure provisioning with Terraform (IaC), creating VMs, VPCs, subnets, and storage as code, which cut environment setup time • Designed and deployed production-grade REST APIs in Node.js that retrieved real-time data from Kafka and fed Grafana dashboards, providing live system metrics that enabled stakeholders to monitor performance instantly Earned AZ-900 certification, demonstrating expertise in Azure fundamental.

• Developed and optimized user interfaces using Bootstrap, HTML5, CSS3, AJAX, and JSON for responsive and dynamic web applications. • Resolved cross-browser compatibility issues and integrated APIs to enhance functionality and user experience.