Available to hire
Hi, I’m Jahnavi Reddy Aell, an AI/ML Engineer with 3+ years of experience building and deploying production-ready ML and NLP systems. I’m proficient in Python, TensorFlow, XGBoost, and cloud platforms like AWS and Azure. I’m currently focusing on Generative AI, agentic AI frameworks, transformers, and LLM integration.
I enjoy collaborating with cross-functional teams to translate complex problems into scalable, ethical AI solutions, prioritizing explainability and robust deployment pipelines.
Skills
Language
English
Advanced
Work Experience
AI/ML Engineer at Mizuho Financial Group
October 1, 2024 - PresentLed development of a predictive credit risk scoring model for underwriting using Python, XGBoost, and AWS SageMaker. Improved default prediction accuracy by 15% and reduced risk exposure by 10%. Engineered features from demographic, transactional, and third-party datasets (SQL/Pandas), boosting precision/recall and raising AUC from 0.78 to 0.86. Implemented cross-validation and hyperparameter tuning to enhance robustness and reduce false positives by 18%. Deployed production-ready model via SageMaker and integrated with loan processing systems through Lambda and S3 to enable real-time scoring, processing 50,000+ applications in 3 months. Identified 3,500+ underserved but creditworthy applicants, contributing to a 12% increase in approved loans and achieving a $45M quarterly lending target. Collaborated with Risk/Compliance to ensure regulatory transparency via SHAP and monitoring dashboards, reducing review cycle from 10 days to 4 days.
AI/ML Engineer at Mass Mutual
May 1, 2021 - July 1, 2023Designed and deployed ML-based fraud detection for insurance claims using Random Forest and XGBoost in Azure ML Studio, increasing fraud detection accuracy by 25% and enabling 75,000+ claim evaluations within 90 days. Engineered features from structured and unstructured data, including medical notes and payment histories, and integrated NLP (TF-IDF, entity extraction) to boost true positives by 2.3x. Deployed in Azure Kubernetes Service with Azure Functions for real-time risk scoring under 1.2 seconds per record. Built automated model monitoring dashboards (Azure Monitor, SHAP) to track performance decay, reducing false positives by 1,800 cases/month and saving ~$600K annually. Collaborated with Risk, Compliance, Claims, and Underwriting to align outputs with audit requirements, shortening investigation cycles by 6 days per claim and recovering/preventing $2.4M in fraudulent payouts over 6 months.
Education
Master of Science in Data Science at Western Michigan University
August 1, 2023 - April 1, 2025Bachelor of Technology in Information Technology at Vignan Institute of Technology and Science
August 1, 2019 - July 1, 2023Qualifications
Oracle Database SQL Certified Associate
January 11, 2030 - December 19, 2025AWS Certified Machine Learning
January 11, 2030 - December 19, 2025OpenAI Prompt Engineering for Developers
January 11, 2030 - December 19, 2025TensorFlow Developer Certificate
January 11, 2030 - December 19, 2025DeepLearning.AI Generative AI Specialization
January 11, 2030 - December 19, 2025Power BI Job Simulation by PwC Switzerland
January 11, 2030 - December 19, 2025Introduction to Generative AI by Google Cloud
January 11, 2030 - December 19, 2025Google AI Essentials by Google
January 11, 2030 - December 19, 2025Industry Experience
Financial Services, Professional Services
Skills
Hire a Data Scientist
We have the best data scientist experts on Twine. Hire a data scientist in Kalamazoo today.