• Own and deliver production ML infrastructure on the Consumer Ads team, including an end-to-end model observability pipeline.

• Implemented evaluation and analysis workflows using SageMaker to assess production models via AUC, precision–recall, confusion matrices, and offline vs online prediction comparisons.

• Optimized ML pipelines for performance and cost, reducing storage by ~89%, memory usage by ~97%, and runtime by ~50%.

• Support on-call rotations and serve as incident commander during production incidents, helping ensure ML system reliability.

• Spearheaded the launch of a new company practice focused on enabling clients to fast-track GenAI and data pipeline adoption on AWS

• Standardized and templatized repeatable delivery patterns, creating internal quickstart frameworks that accelerate production and maintain quality

• Trained and mentored engineers on best practices for delivering production-grade GenAI solutions and modern data architectures

• Acted as AI Solutions Architect in technical discovery and scoping calls, translating client pain points into actionable technical strategies using AWS-native services

• Scoped and delivered solutions for multiple AWS startup clients, including RAG pipelines, data lake integrations, and GenAI-based automation workflows

• Led the development of AI-driven applications using AWS Bedrock and RAG for business automation.

• Designed and optimized backend systems with Django, PostgreSQL, and AWS for high performance.

• Built and maintained Vue.js frontend applications, ensuring seamless user experience.

• Mentored junior developers, conducted workshops, and contributed to setting coding standards and best practices within the engineering team.