AI/ML Engineer at Canada Life

Designed and deployed predictive models for claims risk assessment, reducing fraudulent claim payouts by 15% annually. Built churn prediction models leveraging transaction and engagement data to enable proactive retention strategies that improved customer retention by 11%. Developed and maintained backend applications using Python, NodeJS, and Java; designed RESTful APIs and microservices architectures leveraging AWS cloud technologies for scalable deployment. Implemented NLP models to automate policy document classification and claims text analysis, reducing manual review time by 35%. Implemented personalized recommendation engines for financial products, increasing cross-sell rates by 9%. Partnered with actuarial teams to integrate ML-based mortality and lapse models, improving policy pricing accuracy. Designed and maintained ETL pipelines with PySpark and Airflow for large-scale structured and unstructured insurance data.

AI Development Intern at GoKidu Technologies Corp.

Integrated mobile apps with RESTful APIs, cloud services, and third-party AI/ML APIs (e.g., image recognition, chatbots, recommendation engines). Implemented AI-powered features including NLP, computer vision, and recommendation models. Worked on AI/ML integrations using AWS SageMaker, Bedrock, and Lambda, enhancing system intelligence and automation. Collaborated cross-functionally to explore and apply Generative AI concepts including RAG and LangChain frameworks to improve data-driven decision-making. Applied strong analytical and problem-solving skills in an Agile environment, contributing to innovation, system optimization, and continuous learning. Handled AI data pipelines: managed data collection, formatting, inference calls, and model updates. Conducted debugging, performance optimization, and monitoring of AI-related latency/errors. Managed app deployment and release cycles on Google Play Store and Apple App Store, ensuring compliance with security and privacy guidelines.

Technical Lead (Volunteer) at Voluntary Nature Conservancy

Designed and implemented predictive ML models to analyze participation trends in crowdfunding events, enhancing event planning and marketing strategies. Developed and optimized deep learning-based image classification models for automated spider species identification, improving classification accuracy. Built and fine-tuned CNN models using TensorFlow and PyTorch for object detection and species recognition in biological datasets. Engineered data pipelines for preprocessing and feature extraction, ensuring high-quality input for ML models.

Software Developer AI/ML at Beta Code Technologies Private Limited

Designed and deployed ML models and end-to-end pipelines for real-time data processing using Python, AWS, and SQL, optimizing operational efficiency. Built and optimized microservices and RESTful APIs with FastAPI, integrating ML models seamlessly into production systems. Managed MLOps workflows with MLflow, Airflow, and SageMaker, deploying models in serverless environments (AWS Lambda, GCP Cloud Functions) with CI/CD automation. Integrated model monitoring and experiment tracking using Weights & Biases, enhancing reproducibility and deployment reliability.

ML Engineer (Croc Watch) at Croc Watch

Developed a cross-platform mobile app with Flutter for crocodile population monitoring, featuring offline-first data capture, geotagging, and real-time sync. Integrated Firebase for authentication, storage, and notifications, and built RESTful APIs for centralized data management. Created analytics dashboards to track population trends and habitat patterns, improving conservation insights. Led a small team through code reviews, sprint planning, and CI/CD, collaborating with wildlife experts to refine app usability.